University of Ghana http://ugspace.ug.edu.gh GENETIC ANALYSIS FOR HIGH STARCH, DRY MATTER CONTENT AND STORAGE ROOT YIELD IN CASSAVA (Manihot esculenta Crantz) By KUMBA YANNAH KARIM (10496577) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE AWARD OF DOCTOR OF PHILOSOPHY DEGREE IN PLANT BREEDING WEST AFRICA CENTRE FOR CROP IMPROVEMENT COLLEGE OF BASIC AND APPLIED SCIENCES UNIVERSITY OF GHANA LEGON DECEMBER 2017 University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that except for references to works of other researchers which have been duly cited, this work is my original research and that neither part nor whole has been presented elsewhere for the award of a degree. .................................................. Kumba Yannah Karim Student .................................................. Prof. Eric Y. Danquah Supervisor .................................................. Prof. Essie Blay Supervisor ................................................. Dr.Beatrice E. Ifie Supervisor ……………………………….. Dr.Daniel Dzidzienyo Supervisor ……………………………….. Dr. Jim Whyte Supervisor i University of Ghana http://ugspace.ug.edu.gh ABSTRACT Although cassava (Manihot esculenta Crantz) has benefited from a significant amount of research, there still exist gaps in knowledge on its starch and dry matter content. Studies were conducted to access farmers’ and end-users preferences and also to evaluate improved cassava genotypes for high starch content, high dry matter content and high storage root yield in Sierra Leone. Participatory Rural Appraisal (PRA) was conducted among 360 respondents that comprised farmers, processors, marketers and consumers in 12 chiefdoms in Sierra Leone using focus group discussion and semi-structured interviews. Identified important attributes with cassava production, processing and utilization were high storage root yield, poundability, root taste, dry matter content, resistance to pests and diseases, starch content, and maturity period. These selection criteria reflected the degree of importance attached to multiple needs and priorities, as well as their context in production environments and farming systems. Genetic diversity was assessed among 102 cassava genotypes using a total of 22 agro- morphological traits (11 qualitative and 11 quantitative traits) and 5,600 SNP markers. Significant differences were observed for the quantitative traits. Based on cluster analysis for the qualitative traits, the genotypes were classified into five groups while the dendrogram for the quantitative traits produced four main clusters. The molecular marker based cluster analysis classified the accessions into three main groups. There was a positive correlation between starch percentage and dry matter content. Twenty six genotypes high with dry matter content, starch content and storage root yield were selected and evaluated in three locations for one year to assess genotype by environment interaction effect (GEI) on the expression of dry matter content, starch yield and storage root yield. Analysis of variance indicated significant variation among genotypes, locations and their interactions for fresh storage root yield, dry matter content and starch yield. Fresh storage root yield, dry matter content, starch yield and the other traits evaluated were predominantly under genetic control. Furthermore, twenty four F1 ii University of Ghana http://ugspace.ug.edu.gh families generated from a line x tester mating design revealed a high degree of variation between and within families for all the traits assessed at the seedling stage. Additive gene effects were predominant in the expression of dry matter content and most of the other traits. The physiochemical and functional properties of starch were significantly different within the F1 progenies. The starch content ranging between 61.13 to 69.98% was significant and positively correlated with percentage sugar. Amylopectin was also perfectly correlated with amylose. Individual F1 progenies had higher peak viscosity, set back viscosity and viscosity at breakdown suggesting inherent genetic and biochemical differences among families used in the study. Promising F1 progenies from the seedling stage evaluation would be evaluated in clonal trial, preliminary yield trial and advanced further to multi-location trials where superior genotypes would be selected for improved cassava production in Sierra Leone. iii University of Ghana http://ugspace.ug.edu.gh DEDICATION To my husband Stephen Koroma To my son Stephen Koroma Jr and To my parents– Mr. Henry Kenneth Karim and Mrs. Fatmata Karim iv University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGMENTS I thank the Almighty GOD for his grace, and blessings throughout the period of this study without him this program would have been impossible. To Him I say “Thus Far the Lord has helped me”. I would like to convey my sincere gratitude, appreciation and thanks to various organizations, institutions and individuals who were instrumental in the course of my studies and research. It is not possible to list the names of all individuals, institutions and organizations that contributed or supported this study. I recognize and appreciate your contributions and supports. The ones listed below are some of the many contributors.  WAAPP Sierra Leone for financial supports for me to accomplish my Ph.D. studies.  The Sierra Leone Agricultural Research Institute for granting me study leave.  IDRC Program for financial supports.  Appreciation goes to my supervisory panel of Professor Eric Y. Danquah, Professor Essie Blay, Dr. Daniel Dzidzienyo, and Dr. Beatrice Ifie for their patience, criticism and supervision.  I wish to express my sincere thanks to Dr. Jim Whyte for his good, co-supervision, technical back-stopping on the field work and encouragement given to me for the period of this study—Dr, I’m grateful.  I appreciate the support of Dr. Peter Kulakow, Dr. Ismail Y. Rabbi, Dr. Elizabeth Parkes, Dr. A.G.O. Dixon, Dr. Mrs.B. Maziya Dixon, Dr. Zumanah Bamba, Dr. Alpha Yaya Kamara, Dr. R. Aseidu and Mr. Peter Iluebbey all of IITA Ibadan, Nigeria.  I sincerely appreciate the support I received from the staff of Cassava Breeding Unit, IITA, Ibadan, and Ubiaja stations, Njala Agricultural Research Center Sierra Leone and the entire staff at the Training Unit at IITA. v University of Ghana http://ugspace.ug.edu.gh  Special acknowledgement goes to Dr. Lucky Omoigui who contributed a significant proportion of his valuable knowledge and time, and provided encouragement and assistance when it was needed.  I thank my mother and father who supported me with prayers and love and showed so much concern and took over my son in my absence.  My wonderful siblings (Amie, Abie, Hawa of blessed memory, Sarah, Hannah, Ann, Florence, Kadie, Junior, Abdul, Samuel and Christiana) for your support, love and care for me during the period of this study. Thank you.  Finally, to my lovely husband Stephen I.M.Koroma; you are really appreciated for your love and understanding. You filled the gap for our son during my absence from home, and to my special, amazing son, I appreciate your love, patience and many sacrifices you made for mummy during the study. Stephen Jr. you are wonderful, Mummy loves you.  Many thanks to all and may GOD bless you vi University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS DECLARATION ............................................................................................................................ i ABSTRACT ................................................................................................................................... ii DEDICATION .............................................................................................................................. iv ACKNOWLEDGMENTS ..............................................................................................................v TABLE OF CONTENTS ............................................................................................................. vii LIST OF TABLES ...................................................................................................................... xiv LIST OF FIGURES ................................................................................................................... xvii LIST OF ABBREVIATIONS ..................................................................................................... xix CHAPTER ONE ..............................................................................................................................1 1.0. GENERAL INTRODUCTION .................................................................................................1 CHAPTER TWO .............................................................................................................................5 2.0. LITERATURE REVIEW .........................................................................................................5 2.1. The cassava crop ......................................................................................................................5 2.1.1. Taxonomy of cassava ............................................................................................................5 2.1.2. Flowering ..............................................................................................................................5 2.1.3. Fruit and seed Characteristics ..............................................................................................6 2.2. Selection and evaluation in cassava ..........................................................................................6 2.3.Genetic diversity and variability in cassava ...............................................................................7 2.4. Cassava processing and utilization ...........................................................................................8 2.5. Cassava storage root yield ........................................................................................................9 2.6. Dry matter content in cassava .................................................................................................10 vii University of Ghana http://ugspace.ug.edu.gh 2.7. Factors affecting starch ...........................................................................................................11 2.8. Genetic and phenotypic correlations .......................................................................................11 2.9. Mating design..........................................................................................................................11 2.9.1. Diallel mating design ...........................................................................................................12 2.9.2. Line x Tester ........................................................................................................................13 2.10. Genotype by Environment Interaction ..................................................................................13 CHAPTER THREE .......................................................................................................................15 3.0.CASSAVA FARMERS AND END USER PREFERENCE FOR CASSAVA DRY MATTER CONTENT AND STARCH CONTENT ..............................................................15 3.1. Introduction .............................................................................................................................15 3.2. Materials and methods ............................................................................................................16 3.2.1. Study area.............................................................................................................................16 3.2.2. Sampling procedure and research design .............................................................................18 3.2.3. Data collection ....................................................................................................................20 3.2.4. Data analysis ........................................................................................................................20 3.4. Results ....................................................................................................................................21 3.4.1. Socio-economic characteristics of cassava farmers .............................................................21 3.4.2. Cassava production systems ................................................................................................22 3.4.3. Farmer preference for Starch content in the three study districts ........................................23 3.4.4. Farmers preference for dry matter content in the three studydistricts .................................24 3.4.5. Cassava processors preference for dry matter and starch content in cassava ......................26 3.4.6. Cassava trader’s preferences for starch and dry matter content in cassava products ..........27 3.4.7. Consumer’s preference in cassava starch content and dry matter content ...........................29 3.4.8. Cassava products consumption by districts .........................................................................30 viii University of Ghana http://ugspace.ug.edu.gh 3.4.9. Cassava production systems ................................................................................................31 3.4.10. Farmers selection criteria for cassava cultivars .................................................................32 3.4.11. Processors selection criteria for cassava storage roots ......................................................33 3.4.12. Preferred varieties for different cassava products along the value chain ...........................34 3.4.13. SWOT analysis of cassava farmers....................................................................................35 3.4.14. SWOT analysis of cassava processors ...............................................................................37 3.5. Discussion ..............................................................................................................................39 3.6. Conclusion ..............................................................................................................................41 CHAPTER FOUR ..........................................................................................................................42 4.0. GENETIC DIVERSITY AND VARIABILITY STUDIES OF CASSAVA FOR STARCH CONTENT, DRY MATTER CONTENT AND STORAGE ROOT YIELD ..............................42 4.1. Introduction .............................................................................................................................42 4.2. Materials and methods ............................................................................................................43 4.2.1. Sources of plant material ...................................................................................................43 4.2.2. Experimental design and plot layout ...................................................................................43 4.2.3. Agro-morphological characterization .................................................................................44 4.2.4. Molecular characterization ..................................................................................................46 4.2.4.1. DNA extraction ................................................................................................................46 4.2.4.2. SNP genotyping ...............................................................................................................47 4.2.5. Physicochemical properties determination ..........................................................................47 4.3. Data analysis .........................................................................................................................49 4.4. Results .....................................................................................................................................50 4.4.1. Genetic diversity analysis at the agro - morphological level ..............................................50 ix University of Ghana http://ugspace.ug.edu.gh 4.4.1.1. Frequency distribution of cassava genotypes according to qualitative traits ...................50 4.4.1.2. Analysis of variance for eleven qualitative traits .............................................................54 4.4.1.3. Principal component analysis of qualitative traits ............................................................56 4.4.1.4. Mean values and correlation coefficients for the eleven quantitative traits. ....…………58 4.4.1.5. Representation of variables of quantitative traits ..............................................................61 4.4.2. Hierarchical clustering analysis ..........................................................................................63 4.4.2.1. Dendrogram from qualitative traits ...................................................................................63 4.4.2.2. Dendrogram from quantitative traits .................................................................................66 4.4.2.3. Genetic diversity analysis using SNP markers .................................................................69 4.4.2.4. Clustering analysis ...........................................................................................................73 4.4.3. Determination of cassava starch quality ..............................................................................76 4.4.3.1. Chemical properties of cassava starch fro 96 accessions ..................................................76 4.4.3.2. Pasting properties of cassava starch from 96 accessions ..................................................77 4.4.3.2. Relationships among the physicochemical properties of cassava starch ..........................78 4.5. Discussion ...............................................................................................................................80 4.6.Conclusion ..............................................................................................................................82 CHAPTER FIVE ...........................................................................................................................84 5.0. DETERMINATION OF THE EFFECTS OF GENOTYPE BY ENVIRONMENT INTERACTION OF STARCH CONTENT, DRY MATTER CONTENT AND ROOT YIELD IN CASSAVA. ..............................................................................................................................84 5.1. Introduction .............................................................................................................................84 5.2. Materials and methods ..........................................................................................................85 5.2.1. Experimental sites ..............................................................................................................85 5.2.2. Plant material and experimental design .............................................................................85 x University of Ghana http://ugspace.ug.edu.gh 5.2.3. Data collection and root harvest ........................................................................................86 5.2.4. Data analysis ........................................................................................................................86 5.3. Results .....................................................................................................................................87 5.3.1. Mean performance for Fresh storage Root yield, Starch yield and Dry matter content across the three environments .................................................................................................87 5.3.1.1. Fresh Storage Root Yield .................................................................................................87 5.3.1.2. Starch Yield .....................................................................................................................89 5.3.1.3. Dry matter content ............................................................................................................91 5.3.2. Variation in traits in response to genotypes and locations ...................................................93 5.3.3. Genotype x location interaction effects for fresh storage root yield, starch content and dry matter content a across locations ............................................................................................96 5.3.3.1. GGE biplot of mean and stability performance of 26 cassava genotypes for fresh storage root yield across three environments ..........................................................................96 5.3.3.2. Polygon view of GGE biplot analysis for fresh storage root yield (FSRY) .....................98 5.3.3.3. Ranking of genotypes relative for fresh storage root yield accross the environments .....99 5.3.3.4. GGE biplot of mean and stability performance of 26 cassava genotypes for Starch yield over two environments …………………………………………………………………….100 5.3.3.5. Polygon view of GGE biplot analysis of MET data for starch yield (SY) .....................102 5.3.3.6. Ranking of genotypes relative for starch yield accross the environment ......................104 5.3.3.7. GGE biplot of mean and stability performance of 26 cassava genotypes for dry matter content across three environments ........................................................................................105 5.3.3.8. Polygon view of GGE biplot analysis of MET data for dry matter content (DM) .........106 5.3.3.9. Ranking of genotypes relative to highest starch yielding environment ..........................107 5.3.3.10. Phenotypic correlations among yield related traits .......................................................108 5.4. Discussion .............................................................................................................................108 xi University of Ghana http://ugspace.ug.edu.gh 5.5. Conclusion ........................................................................................................................ ...111 CHAPTER SIX ............................................................................................................................112 6.0. EVALUTION OF F1 SEEDLING POPULATIONS FOR DRY MATTER CONTENT AND STARCH CHEMICAL PROPERTIES IN CASSAVA ....................................................112 6.1. Introduction .........................................................................................................................112 6.2. Materials and methods ........................................................................................................113 6.2.1. Experimental site ............................................................................................................113 6.2.2. Germplasm source and parental selection and hybridization ..........................................113 6.2.3. Experiment location ........................................................................................................114 6.2.4. Trial design .......................................................................................................................115 6.2.5. Data collection ...............................................................................................................115 6.2.6. Data analysis .....................................................................................................................116 6.3. Result ...................................................................................................................................117 6. 3.1. Performance of the individual genotypes within families ...............................................117 6. 3.2.Analysis of variance using line x tester design .................................................................118 6. 3.3. Mean performances of the 24 F1 families ........................................................................120 6.3.4. General combining ability..................................................................................................122 6.3.5. Specific combining ability ................................................................................................125 6.3.6.Contribution of traits to the families’ variability ...............................................................127 6.3.7. Phenotypic correlations for dry matter content and agronomic traits ................................128 6.3.8. Physiochemical characteristics of cassava F1 progenies ...................................................129 6.3.8.1. Chemical characteristics of F1 progenies in cassava starch ............................................129 6.3.8.2. Pasting properties of cassava starch ................................................................................131 6.3.8.4. Correlations among the physicochemical and functional properties of cassava starch ..133 xii University of Ghana http://ugspace.ug.edu.gh 6.4. Discussion ............................................................................................................................134 6.5. Conclusion ............................................................................................................................137 CHAPTER SEVEN .....................................................................................................................138 7.0. Conclusions and Recommendations .....................................................................................138 7.1. Conclusions ...........................................................................................................................138 7.2. Recommendations .................................................................................................................141 BIBLIOGRAPHY ........................................................................................................................142 APPENDICES ............................................................................................................................165 xiii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 3.1: Sampling Procedure and Research Design ................................................................ 19 Table 3.2: Percentage distributions of cassava farmers across the three study district .............. 21 Table 3.3: Percentage distribution of farmer’s preference for starch content in cassava ........... 24 Table 3.4: Percentage distribution of farmers preference for dry matter in cassava ................. 25 Table 3.5: Percentage distribution of processors preference for cassava roots with starch content and dry matter ................................................................................................................ 27 Table 3.6:Percentage distribution of trader’s preferences for starch and dry matter in cassava products in the three study districts ............................................................................................ 28 Table 3.7: Percentage distribution of consumer’s preference for starch and dry matter in cassava products across the three study districts ........................................................................ 30 Table 3.8: Cassava traits ranking by farmers across the three study districts ............................. 33 Table 3.9: Cassava traits ranking by processors across the three study districts ........................ 34 Table 3.10: Preferred cassava varieties for different cassava products in the three study districts ....................................................................................................................................... 35 Table 3.11: SWOT analysis for cassava farmers in the three study districts .............................. 36 Table 3.12: SWOT analysis for cassava processors in the three study districts ...................... 38 Table 4.1 : Description of 102 cassava genotypes used for the study ....................................... 45 Table 4.2: Qualitative and quantitative traits used for characterize the cassava genotypes ...... 46 Table 4.3: Probability values, means and coefficient of variation of qualitative traits of 102 cassava genotypes ..................................................................................................................... 55 Table 4.4:Principal component analysis, eigenvalues and percentage variation of eleven qualitative traits of 102 cassava genotypes .................................................................................................. 57 xiv University of Ghana http://ugspace.ug.edu.gh Table 4.5: Probability values,means and coefficient of variationof quantitative traits of 102 cassava genotypes ...................................................................................................................... 59 Table 4.6: Correlation coefficients among 11quantitative traits inof 102 cassava genotypes ... 60 Table 4.7: Principal component analysis, eigenvalues and percentage variation of eleven quantitative traits of 102 cassava genotypes .............................................................................. 62 Table 4.8: Grouping for 102 genotypes based on qualitative traits ........................................... 65 Table 4.9: Grouping for 102 genotypes based on quantitative traits ......................................... 68 Table 4.10: Summary statistics of genetic variation using 5,600 SNP markers among 102 cassava accessions ..................................................................................................................... 70 Table 4.10: (cont’d) summary statistics of genetic variation using 5,600 SNP markers among cassava accessions ..................................................................................................................... 71 Table 4.10:(cont’d) summary statistics of genetic variation using 5,600 SNP markers among cassava accessions ...................................................................................................................... 72 Table 4.11: Cluster groupings of the cassava accessions based on SNP markers ..................... 75 Table 4.12: Summary statistics of starch quality traits from 102 cassava genotypes ................ 76 Table 4.13:Summary statistics for pasting properties from starches of 102 cassava genotypes 77 Table 4.14: Pearson correlations among the physicochemical and functional properties of cassava starch ............................................................................................................................. 79 Table 5.1:Performance of the 26 cassava genotypes evaluated for fresh storage root yield across three locations ........................................................................................................................... 88 Table 5.2: Performance of the 26 cassava genotypes evaluated for starch yield across three locations ..................................................................................................................................... 90 xv University of Ghana http://ugspace.ug.edu.gh Table 5.3: Performance of the 26 cassava genotypes evaluated for dry matter content across three locations ............................................................................................................................ 92 Table 5.4: Mean squares of combined analysis of variance showing the reaction of the 26 cassava genotypes evaluated for dry matter content and its components in three locations ..... 94 Table 5.5: Mean squares of AMMI analysis of variance showing the reaction of the 26 cassava genotypes evaluated for dry matter content and its components in three locations .................... 95 Table 5.6: Phenotypic correlations among yield and it related traits ........................................ 108 Table 6.1:Description of parent genotypes used in the line x tester mating design ................. 114 Table 6.2: Statistics summary of 8 traits evaluated for F1 seedling ......................................... 117 Table 6.3: Line x Tester analysis of 24 families evaluated at seedling stage ......................... 119 Table 6.4: Performances of 24 families at the seedling stage .................................................. 121 Table 6.5: General Combined ability for six lines (Female) .................................................... 123 Table 6.6: General combined ability for four testers (male) ..................................................... 124 Table 6.7: Specific combining ability effects for 8 traits from the line x tester analysis of 24 families evaluated at 12 months after planting at Njala 2016/17 .............................................. 126 Table 6.8: Principal component loadings, eigenvalues and percentage variation of eight traits evaluated in 24 F1 cassava seedling stage families at 12 months after planting, ..................... 128 Table 6.9: Phenotypic correlation coefficients for dry matter content and agronomic traits for 24 families evaluated at the seedling stage .............................................................................. 129 Table 6.10: Proximate analysis for physiochemical characteristics of the F1 progenies starch ................................................................................................................................................... 130 Table 6.11:Proximate analysis for pasting properties of cassava starches .............................. 132 Table 6.12: Correlation matrix between the different starch parameters ................................. 133 xvi University of Ghana http://ugspace.ug.edu.gh LIST OF FIGURES Figure 3.1: Map of Sierra Leone showing the three study districts: Bombali, Kambia and Bo. ..................................................................................................................................................... 17 Figure 3.2: Percentage distribution of type cassava growers in the three study districts ........... 22 Figure 3.3: Percentage distribution of respondent to Cassava products consumer across the three study district ....................................................................................................................... 31 Figure 3.4: Percentage distribution of type of cassava growers and processors across the three study districts .............................................................................................................................. 32 Figure 4.1: Percentage distribution of six cassava qualitative traits ........................................... 52 Figure 4.2: Percentage distribution of five cassava qualitative traits ......................................... 53 Figure 4.3: Ward method of classification of genotypes based on eleven qualitative traits ...... 64 Figure 4.4: Ward method of classification of genotypes based on eleven qualitative traits. ...... 67 Figure 4.5: Dendrogram of 96 cassava accessions based on SNP markers ................................ 74 Figure 5.1: GGE biplot of mean and stability performance of 26 cassava genotypes for yield over three environments ............................................................................................................. 97 Figure 5.2:GGE biplot for best genotypes for fresh storage yield across different environments ..................................................................................................................................................... 98 Figure 5.3: The average-environment coordination (AEC) view of genotypes relative to an ideal environment for fresh storage root yield ............................................................................ 99 Figure 5.4: GGE biplot of mean and stability performance of 26 cassava genotypes for starch yield across three environments ............................................................................................... 101 Figure 5.5: GGE biplot for best genotypes for starch yield across three different environments ................................................................................................................................................... 103 xvii University of Ghana http://ugspace.ug.edu.gh Figure 5.6: The average-environment coordination (AEC) view of genotypes relative to an ideal environment for starch yield ............................................................................................ 104 Figure 5.7: The GGE biplot showing mean performance and stability of 26 cassava genotypes for dry matter content................................................................................................................ 105 Figure 5.8: GGE biplot for best genotypes in different environments for dry matter content . 106 Figure 5.9: The average-environment coordination (AEC) view of genotypes relative to an ideal environment for dry matter content.................................................................................. 107 xviii University of Ghana http://ugspace.ug.edu.gh LIST OF ABBREVIATIONS ACMV African Cassava Mosaic Virus AFLP Amplified Fragment Length Polymorphism AGPase Glucose Pyrophosphorylase CBB Cassava Bacterial Blight CBSD Cassava Brown Streak Disease CBSV Cassava Brown Steak Virus CIAT International Centre for Tropical Agriculture CMD Cassava Mosaic Disease COSCA Collaborative Study of Cassava in Africa DAP Date after planting DM Dry Matter DNA Deoxyribonucleic Acid ESTs Expressed Sequence Tags FAO Food and Agriculture Organization FGD Focus Group Discussion GBS Genotyping by sequencing GEI Genotype and Environment Interaction GCA General Combining Ability IITA International Institute of Tropical Agriculture MAF Minor allele frequency MAP Month after planting MAS Marker Assisted Selection PCA Principal Component Analysis PIC Polymorphic Information Content PRA Participatory Rural Appraisal xix University of Ghana http://ugspace.ug.edu.gh RAPD Random Amplified Polymorphic DNA RFLP Restriction Fragment Length Polymorphism SAS Statistical Analysis System SCA Specific Combining Ability SCARs Sequence Characterized Amplified regions SNP Single Nucleotide Polymorphism SSR Simple Sequence Repeats SWOT Strengths Weakness Opportunities Threats UNIDO United Nation Development organization xx University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE 1.0. GENERAL INTRODUCTION Cassava (Manihot esculenta Crantz) is a very important root crop, containing high carbohydrate levels and used for human consumption, animal feed and industrial applications (Sánchez et al., 2009). It is extensively cultivated as an annual crop in tropical and subtropical region for its edible storage root as major source of carbohydrate. The cassava crop is essentially grown between latitudes 30° N and 30° S (Zira, 2007). Cassava tolerates hot climate, but a critical point exists between a daily average temperature of 18 and 20°C, below which the plants will not grow normally and the yields decrease rapidly (Zira, 2007). The crop is a high starch producer with levels of up to 90% of its total storage root dry mass (Jansson et al., 2009), and about 700 million people derive more than 500 kcal day-1 from consuming cassava storage roots (Kawano et al., 2003; Burns et al., 2011). Cassava is currently the world’s fourth most important staple and carbohydrate rich food crop (El-Sharkawy, 2012), with a world wide production estimated at 257 million tonnes (MT), of which about 146 MT comes from Africa (FAO, 2012). Cassava plays a number of different but equally important roles in African development, including being a rural as well as urban food staple an industrial raw material and as livestock feed. According to Scott et al. (2000), the production of cassava is expected to increase to 290 million metric tons by 2020. The annual per capital consumption of cassava in sub-Saharan Africa is estimated at 106 kg and is reported to have increased by 2.1% annually between 1983 and 1996 (Scott et al., 2000).The crop is highly suited to intercropping with many types of crops and its time of harvest is flexible and has a wide variety of food, feed and industrial uses (Westby, 2002; Jansson et al., 2009). These attributes make cassava a significant crop in food production and income generation, particularly benefitting the poor in the tropical regions of the world (Henry and 1 University of Ghana http://ugspace.ug.edu.gh Hershey, 2002). However, the bulk of cassava production is consumed as food (Nweke et al., 2002). The average yield of cassava world wide is only 12– 13 t/ha, but its potential yield under optimal conditions is almost seven times higher (80 t/ha) (FAO, 2013). Cassava is processed into common based products including: raw tuber, gari, starch, cassava bread and boiled cassava with beans, all of which are traded in Sierra Leone (Latif et al., 2009). The leaves are used to prepare a very popular national cassava leaf sauce (Sahr et al., 2012). Despite its importance and tremendous contribution, production is faced with several constraints in Sierra Leone which include; usage of low yielding varieties, untimely and inadequate weeding, low and unsteady rainfall, soil infertility, sloping grounds, improper planting, handling and storage of planting materials, pests and diseases such as cassava mosaic, cassava blight, cassava anthracnose and weed infestation. Small holder production and use often faced with other problems, including unreliability of supply, uneven quality of products, low prices and costly marketing activities associated with development of cassava products and markets. In spite of the efforts to introduce improved cassava varieties, the yield in Sierra Leone is still low compared to other sub-Sahara African countries (Jalloh, 2000) estimated at 7.18 t/ha according to (FAO 2012). Every part of a cassava plant can be utilised, but its starchy storage roots are by far the most commonly used (Ceballos et al., 2004; Ojulong et al., 2007). The storage roots are rich source of carbohydrates (Westby, 2002; Jansson et al., 2009; El-Sharkawy, 2012) mostly present as starch (31% of fresh mass), with smaller amounts of free sugars (less than 1% of fresh mass). Starch is a valuable ingredient for the food industry, it is widely used as a thickener, gelling, bulking and water retention agents (Niba et al., 2001; Singh et al., 2003). Cassava starch is also used as raw material for the production of paper, textiles, monosodium glutamate (MSG), and as an important flavouring agent in Asian cooking (FAO, 2001).The 2 University of Ghana http://ugspace.ug.edu.gh physicochemical and functional properties of starch are the most important characteristics in their many applications, and have a large impact on product quality (Nuwamanya et al.,2010). Dry matter content is an important determinant of storage root yield in cassava and could be an important selection criterion in breeding programmes to enhance yield (Mohamed et al., 2009). The dry matter content which is also referred to as the dry weight is however controlled by polygenic additive factors (IITA,1985; Kawano et al.,987). According to Lian (1985) the dry matter is influenced by several factors such as the age of the plant, crop season, location and efficiency of the canopy to trap sunlight. Dry matter of cassava varies from one accession to another and ranges between 17% and 47% with the majority lying between 20% and 40% (Barima et al., 2000); values above 30% are considered high. High dry matter content and starch are the main characteristics preferred by consumers and processors of cassava, for instance in sub-Saharan Africa, small-scale farmers prefer cassava varieties that have high dry matter content and starch (Mwanga et al., 2007;Cervantes-Flores et al., 2010). However, cassava breeding for high dry matter content is very challenging due to its genetic complexity, and marker-assisted breeding tools are needed to facilitate crop improvement. In spite of its global economic importance, cassava has traditionally received less research attention than temperate crops (Cock, 1985). Therefore, selecting for desirable traits is required to enhance cassava breeding (Asante and Dixon, 2006). The primary objective of this research was to evaluate improved cassava genotypes for high starch content, high dry matter and storage root yield, for adoption by farmers and end-user in Sierra Leone. The specific research objectives were to: a. Assess end user preferences for dry matter content and starch along the cassava value chain in Sierra Leone; 3 University of Ghana http://ugspace.ug.edu.gh b. Estimate the genetic diversity and variability for starch, root dry matter content and root yield of cassava genotypes; c. Estimate the effects of genotype by environment interaction on fresh storage root yield, dry matter content and starch yield in cassava; d. Evaluate cassava F1 families for dry matter content and starch physiological properties. 4 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO 2.0. LITERATURE REVIEW 2.1. The cassava crop 2.1.1. Taxonomy of cassava Cassava (Manihot esculenta Crantz) is a diploid (2n = 36) crop and belong to the spurge family Euphorbiaceae. The genus of cassava has two sections, the Arborae, containing tree species, and the Fructicosae, containing the shrubs (Otim-Nape et al., 2001; Jennings and Iglesias, 2002). It is believed to be a segmental allotetraploid because chromosomes at metaphase one and anaphase show a high number of duplicated nucleolar chromosomes (Kawano, 1980). There are no genetic and cytological barriers in the species of the Manihot genus (Nassar, 2002), thus crosses can be made between species in the genus. Cassava is widely cultivated and has been classified according to different morphological traits. Cyanogenic glycoside content has been used by farmers to classify cassava cultivars in two groups as sweet when cyanogenic glycoside content is low, and bitter when cyanogenic glycoside content is high (Chiwona-Karltum et al., 2004). 2.1.2. Flowering Most cassava plants are monoecious and a few are dioecious, which make them out-crossers. In the monoecious species, they are protogenoius that is the pistillate flowers open before staminate flowers of the same inflorescence. The flowers of cassava plant are small, with the male flower being about 0.5 cm in diameter, and the female flower slightly larger. The inflorescence of the female flowers open first and the male flowers open one in a few weeks later. According to (Byrne, 1984), the flowering of the plant depends on the specific genotype and environmental conditions, and varied from 1 to 24 months. Flowers of cassava plant usually open around mid-day, and remain open for one day (Ceballos et al., 2002). 5 University of Ghana http://ugspace.ug.edu.gh Cassava is classified into different growing season and environments; so that breeders may take account of the flowering habits of the plants for cross (Ceballos et al., 2002). For some varieties, flowering induction appears to depend on long photoperiods – up to 16-hour day length, associated with temperatures of around 24ºC (Alves, 2002). Cassava is predominantly cross pollinated, but self-pollination has been reported (Jenningsand Iglesias, 2002; Nassar, 2002). 2.1.3. Fruit and seed characteristics The seed of cassava is viable two months after pollination and the fruit of cassava generally matures between 75 to 90 days after pollination (Alves, 2002). The fruit is a trilocular schizocarp, and seeds are ovoid-ellipsoidal, approximately 100 mm long and 4 to 6 mm thick (Alves, 2002). Cassava seed weight ranges from 95 to 135 mg seed-1(Alves, 2002; Rajendran et al., 2005). Seed production and viability are variable in cassava, depending largely on the quality of the female parent (Kawano, 1980). Cassava seeds are recalcitrant; studies have shown that after six months of storage at laboratory temperature and between 5.9% and 1.9% moisture content, germination reduced from 80 to 28% (Ellis et al., 1981). Newly harvested seeds are dormant, requiring 3 to 6 months of storage before they will germinate (Jennings and Iglesias, 2002). Seeds can remain viable for up to 1 year, although germination percentages may decline substantially after 6 months (Rajendran et al., 2000). Cassava genotype is extremely heterogeneous, and propagation from sexual seed results in wide and unpredictable diversity of phenotypes, which is of interest to breeders for propagation (Ceballos et al., 2004). 2.2. Selection and evaluation in cassava Accord to Hahn et al. (1979) any breeding programme should have priority research themes and objectives that are clearly established on the basis of the production constraints to be 6 University of Ghana http://ugspace.ug.edu.gh resolved. Relatively little attention has been to the genetic improvement of root dry matter and starch content. Intensive selection for disease resistance and root yield potential have restricted the availability of genetic variability for dry matter content and starch (Iglesias and Hershey, 1994). Early generation testing is used in self- and cross-pollinated species to estimate the genetic potential of an individual (Fehr, 1987). The early selection in cassava includes seedling and clonal (single row) stages are based on high heritability traits such as plant type,branching habits, harvest index and diseases (Hahn et al., 1980; Hershey, 1988; Iglesias et al., 1994 and Kawano, 1990). The selected seedlings are uprooted after 12 months and screened for conformation and root characteristics. Kawano et al. (1982) suggested that it is important to establish seedling populations at low planting densities to give all plants an opportunity to express their genetic capacity and to minimize the effects of intergenotypic competition. Ceballos et al., (2004) also reported that, large number of materials at lower costs visual evaluation with few data recording has been a common feature in the first stages of selection. Byrne (1984) observed that there were significant correlations (r=0.48**) between dry matter content in seedling and single row trials, and suggested that evaluation for dry matter content was feasible at the F1 stage. 2.3. Genetic diversity and variability in cassava Genetic diversity and variability studies by plant breeders, geneticists, and taxonomists are of great importance (Prince et al., 1995). Genetic diversity is the first step to understand the genetic variability in the germplasm (Hurtado et al., 2008). Genetic distances within a population provides for an effective parental selection during sampling of genotypes (Meredith and Bridge, 1984). The methods investigate the genetic variability of cassava by the use of DNA-based markers which have contributed to the understanding of different cassava breeding and genetic analysis studied (Olsen and Schaal, 2001) and facilitates the improvement of the genetic diversity ( Second et al., 1997; Okai, 2001; Elias et al., 2001; 7 University of Ghana http://ugspace.ug.edu.gh Mkumbira et al., 2003). Studies on cassava based on DNA sequence, SNP marker and SSR marker have revealed genetic variation in cassava (Olsen and Schaal, 2001). Therefore, genetic variability within germplasm can be assessed using agro-morphological descriptors and molecular markers 2.4. Cassava processing and utilization Utilization of cassava has advanced, from subsistence and household consumption to the processing of industrialized commodities and domestic/human food usage account for the greater percentage of the crop produced in Africa and Asia (Westby, 2002; UNIDO, 2006). Cassava is currently grown by smallholders farmers for use as food and as cash crop in Africa. Storage root requirements for the fresh market include taste, the size of the root and low levels of cyanogenic potential in roots (Kawano et al., 1998). For the processing market, cultivars with higher root yield, high starch content and dry matter content are required (Kawano et al., 1998). Prominent among the characteristics are high yield, early bulking and high dry matter content (Nweke et al., 1998). Similarly, Temu et al. (2002) study of cassava markets in Tanzania revealed three key attributes that lead to variety acceptance by consumers: high dry matter content, low fiber content and sweetness. However, Hillocks, (2002) reported that, farmers normally select desired characteristics over a period of time; cultivars with undesirable characteristics are abandoned and the most frequent reason given by farmers for discarding varieties was late bulking. Similarly, a high yielding improved cassava variety could be rejected simply because when processed into flour and cooked into stiff porridge it becomes watery. Therefore, there is a need to include root quality aspects such as starch characteristics in breeding programmes to enhance the adoption of improved varieties. Cassava storage roots essentially contain large carbohydrate reserves, mainly of starch; therefore, cultivars with 8 University of Ghana http://ugspace.ug.edu.gh high dry matter content are important (Tan and Mak, 1995). In addition, high dry matter content is important because it ensures a high recovery rate of dried roots (Byrne, 1984). Graham et al. (1999) commented that for the vast majority of nine uses, cultivars of high dry matter content are mostly preferred and participatory plant breeding is important to capture and include desired characteristics by farmers in the breeding programme. 2.5. Cassava storage root yield Storage root in plant is the main goal of any breeding programme. It is a complex quantitatively inherited trait and difficult to improve directly. Although, progress has been made in terms of storage root yields (Kawano et al., 1987), the problem of identifying appropriate indicators of yield during selection process still remain a challenge. Hahn et al. (1979) suggested that yield in cassava has three major components; i) the number of storage root per unit area, ii) the average root weight and iii) the percentage of dry matter content of storage roots. However, attempts to identify the index of selection using simple correction analysis did not confirm dry matter content to be important indication of the storage root yield (Ntawuruhunga, 1992). Kang (1994), also suggested that improvement of such complex traits could be handled through indirect selection that is, selection for a component trait or trait involved in the pathway leading to the formation of the complex trait. Fresh storage root yield of cassava is multiplied by the root dry matter percentage which constitutes the dry matter of cassava (Kawano et al., 1987). The cassava roots are expressed in fresh root weight, but there are significant varietal differences for root dry matter content (CIAT, 1976). Kawano et al. (1987) reported that, there was no indication of a negative correlation between fresh yield and dry matter content.. According to Adepoju et al.(2010), the value of food in cassava is greatly compromised by its toxic hydrogen cyanide content. A large amount of variation exists among the cassava leaf, 9 University of Ghana http://ugspace.ug.edu.gh stem and root characteristics. These characteristics, which include leaf morphology, stem colour, branching habit and storage root shape and colour, may influence cassava yield (Ntawuruhunga and Dixon, 2010). 2.6. Dry matter content in cassava Dry matter content is an important determinant of storage root yield in cassava (Mohamed et al., 2009). According to Lian (1985), dry matter content is influenced by several factors such as the age of the plant, season of the crop, location and efficiency of the canopy to trap sunlight. Dry matter content in cassava roots vary from 15 to 45% depending on the age of the crop, the type of genotype used and environmental conditions (Graham et al., 1999; Babayoko et al., 2009; Okechuku and Dixon, 2009). Tan and Mak,(1995) reported that, high dry matter content is important especially when roots are used as food, feed and industrial raw materials. Alves (2002) also reported that, dry matter accumulation depends on the availability of photo-assimilates and the sink capacity of the storage roots. There are two methods for determination of the dry matter content in cassava. According to (Jennings and Iglesias, 2002),the specific gravity method is a quick method for determining root dry matter content. Normally, unpeeled fresh roots are weighed in air and then in water. The other method is the forced oven dry method (Wholey and Booth, 1979; Kawano et al., 1987; Jennings and Iglesias, 2002). The fresh tuber is weighed, dried in an oven for 24 hours o at 105 C and reweighed to determine the percentage dry matter content. Inheritance of dry matter content as an agronomic trait has been undertaken in cassava (Ceballos et al., 2004; Jaramillo et al., 2005). However, few studies have been conducted on genetic analyses suggesting that inheritance of root dry matter content is controlled by polygenic additive factors (Kawano, 1987). 10 University of Ghana http://ugspace.ug.edu.gh 2.7. Factors affecting starch According to Chatakanonta et al. (2003), delay in the harvesting of cassava roots until 14-16 months results in increased fiber content, increased starch and decreased water content. Starch functionality in cassava depends on the environmental conditions and the age of the crop (Asaoka et al., 1992.). Santisopasri et al. (1998) studying cassava varieties in Thailand, observed a high content of starch between 26-28% . They also observed that roots with a high amount of biochemicals like lipid, protein, cyanide, phenolic compounds and fiber had lower of starch quality. Howeler (2002) also reported that, potassium (K) fertilizer increased root yield and starch content. A similar result of an increase in starch content with increasing application of Potassium (K) have been reported at CIAT as well as in Southern Vietnam (Nguyen et al., 1998; CIAT, 1982). 2.8. Genetic and phenotypic correlations in cassava Correlation coefficients are useful for indirect selection in cassava; however, they depend on estimation of heritability and the genetic correlation (Kawano et al., 1998). Kawano et al. (1998) reported that, direct selection for yield was less effective than indirect selection through harvest index, because of the significant differences between the yield performances of the same genotype. Genetic correlations are important to determine the degree of association between traits and enhancing selection (Falconer, 1989; Hallauer and Miranda 1988). Root dry matter content and starch have been reported to have a high and significant correlation (r = 0.81; IITA, 1974; CIAT, 1975) hence there is a possibility of employing indirect selection to improve starch content. 2.9. Mating designs According to Nduwumuremyi et al. (2013) mating design is a process used by breeders and geneticists in various ways and arrangement to generate improved plants or varieties. 11 University of Ghana http://ugspace.ug.edu.gh However, additive genetic variation is more important at the early stage of breeding (Sprague and Tatum, 1942). Non-additive effects become more important when selection proceeds because the selected material has greater similarity, thereby largely eliminating additive effects. The particular mating design which reflects these processes include: biparental progenies (BIP), North Carolina I (NCI) (Nested design), North Carolina II (NCII), North Carolina III (NCIII) and diallel (I, II, III and IV), Line x tester design, Polycross and Topcross. However, there are factors affecting choice of mating design which include, (I) the type of pollination (self- or cross-pollinated), (II) the type of crossing to be used (artificial or natural) (III), the type of pollen dissemination (wind or insect), (IV) the presence of a male - sterility system and (V) the purpose of the project (for breeding or genetic studies). 2.9.1.Diallel mating design Diallel mating design allows an estimation of heritability components . Hallauer and Miranda (1988) commented that extensive theoretical research and discussion have been presented from the interpretations and inference that can be obtained from analysis of the diallel crosses. Diallel analysis, although effective and widely used, does not provide estimates of non-allelic interactions (Sharma and Sain, 2004; Hill et al., 1998). Significant epistatic variation clearly indicates the role of epistatic gene actions besides additive and dominance gene actions, which play a major role in the expression of heterotic potential (Stuber and Moll, 1974; Brim and Cocker ham, 1961; Hayman, 1958). Four methods of diallel crossing includes: 1) full diallel, the parents, F1 and reciprocals included; 2) half diallel, parents and F1’s included, but no reciprocals; 3) F1’s and reciprocals included, but no parents and 4) F1’s included, excluding reciprocals or parents. 12 University of Ghana http://ugspace.ug.edu.gh 2.9.2. Line x Tester Line by tester (L × T) mating design , proposed by Kempthorne et al. (1957), is an extension of top cross design (Sharma, 2006) in that instead of using one tester as in top cross, more than one tester is used. The design provides SCA of each cross and GCA of both the lines and of the testers. Both the lines and testers have different sets of genotypes (Farhan et al., 2012). In line by tester mating design, testers that can be used in a breeding program may either be genetically narrow or broad-based, related to the lines being evaluated for favorable alleles and yield in plant breeding (Ali et al., 2011; Packer, 2007). According to Packer (2007) effective tester should rank inbred lines for performance in hybrid combination, and should maximize the variance between test cross progeny to allow effective discrimination of new inbred lines. Line by tester studies have been conducted by several researchers (Gouda et al., 2013) with different findings on general and specific combining abilities for yield and its related traits. Shushay et al. (2013) working on line by tester analysis of inbred lines in maize for yield related traits reported highly significant SCA for grain yield. 2.10. Genotype by environment interaction Genotype by environment interaction refers to large differential genotypic responces under different environmental condition. The success of cassava as a cash crop and food security crop largely depends on its ability to yield well under poor management cropping system and grow where other crops would fail. Recent studies on genotype by environment interactions in some economic crops include the work by Akinyele and Osekita (2011), Sakin et al.(2011), Ngeve et al.(2005) and Kilic et al.(2009). According to Ssemakula and Dixon.(2007), large G x E interaction usually impairs the accuracy of yield estimation and reduces the relationship between genotypic and phenotypic values. Mather and Jinks (1982) reported on two type of genotype by environment 13 University of Ghana http://ugspace.ug.edu.gh interactions: (1) micro-environmental which cannot be easily predicted and (2) macro- environmental variances which can be easily predicted. Various studies using GEI have revealed genetic variation among the type of genotypes evaluated and the type of environment (Lebot, 2009). 14 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 3.0 END USER PREFERENCES FOR CASSAVA DRY MATTER CONTENT AND STARCH CONTENT 3.1. Introduction Starch constitutes the main component of the cassava root (Ceballos et al., 2006) and thus plays an important role in the use of cassava as a food and industrial crop. The economic value for cassava products for end users emanates from the dry matter content and starch content. The performance of cassava starch and dry matter in food, feed and other industrial applications vary according to variety from which the product was obtained (Benesi et al., 2003; Singh et al., 2005). In Sierra Leone, several cultivars are grown some of which are known to farmers as having high dry matter content and starch content since most of the cassava produced and consumed are mainly processed as gari, fufu and tho. According to Okogbenin and Fregene (2002), dry matter content and starch are important requirements for the transition of cassava from a traditional to an industrial crop. In some African countries, Participatory Rural Appraisal (PRA) tools have been used to determine end users’ preferences in many crop varieties.Manu-Aduening et al. (2007) used participatory rural appraisal (PRA) to describe the characteristics needed for cassava varieties in Ghana and reported that farmers preferred cassava varieties that have early growth and vigour to suppress weeds, early maturing, high yield, good cooking quality for making fufu and suitability for intercropping. Participatory plant breeding approaches such as surveys and focus group discussions have been deemed necessary to elicit such vital information on what is needed by farmers (Ceccarelli and Grando, 2007; Efisue et al., 2008 and Kamau et al., 2011; Parkes, 2011; Were et al., 2012).With the increase in commercialization of cassava and 15 University of Ghana http://ugspace.ug.edu.gh cassava products, the demand for higher quality varieties with high dry matter and starch content that meet various needs will become the next obvious challenge This study was designed to assess end users’ preferences for dry matter and starch content along the cassava value chain in Sierra Leone. Specific objectives were to: 1. Assess stakeholder preferences for dry matter content and starch content 2. Decipher difference along the cassava value chain in the production of storage roots with dry matter content and starch content and utilization in cassava based products. 3.2. Materials and methods 3.2.1 Study area The PRA study was conducted in the Northern, Eastern and Southern regions located in the coastal belt agro-ecological zones of Sierra Leone (Figure 3.1). The climate is tropical, hot all year round, with a dry season and a rainy season, due to the African monsoon, which runs from May to November in the north, from April to November in the east, and May to October in the south. The annual rainfall varies from 2,000 to 3,000 mm, with a the rainfall in the coastal area. 16 University of Ghana http://ugspace.ug.edu.gh Fig. 3.1 Map of Sierra Leone showing the three study districts: Bombali, Kambia and Bo Source:https://www.mapsofworld.com/sierra-leone/ 17 University of Ghana http://ugspace.ug.edu.gh 3.2.2. Sampling procedure and research design A purposive sampling procedure was used to identify region, villages, farmers, processors, traders and consumers. Three regions, Northern, Eastern and Southern were selected for this study. The regions were selected based on their potential for cassava production. Farmers, processors, traders and consumers that were involved in the formal and informal interviews were randomly selected from the village with the help of village and village hamlet leaders and Agricultural Extension Officers. Four chiefdoms were selected from each district. Four research designs were selected for the study: a. A mixed-method research design that combines qualitative and quantitative approaches was applied. b. Qualitative Approach: Focus group discussions c. Quantitative Approach: Individual interviews with structured questionnaires d. A multistage sampling was used to arrive at 360 sampling units that were then distributed among producers, processors, traders and consumers. The sampling method was determined as: z2 pq n  d 2 Where:  n= sample size  Z = 1.96  P = population  q = is a weighting variable computed as 1 – p  d = margin of error 18 University of Ghana http://ugspace.ug.edu.gh One (1) community per chiefdom was randomly selected for the focus group discussion. Table 3.1 represents the summary of sample procedure and research design in the study areas Table 3.1: Sampling Procedure and Research Design Stage Category Sampling method Sample size Stage: 1 3 regions (East, north and South) Purposive 360 actors Stage: 2 3 districts (Bo, Bombali and Kenema) Simple random 120 actors per district Stage 3 12 chiefdoms Simple random 30 actors per chiefdom Stage 4 12 Chiefdoms Producers Systematic 10 Producers per chiefdom Consumers systematic 6 Consumers per chiefdom Traders Simple random 8 Traders per chiefdom Processors Simple random 6 Processors per chiefdom 19 University of Ghana http://ugspace.ug.edu.gh 3.2.3. Data collection Data collection was based on focus group discussion and questionnaires administration. In each village, a focus group discussion was conducted with groups of 12 representatives. A checklist was prepared to guide the discussion. Ranking technique was employed to complement semi-structured interviews. A total of 360 questionnaires were administered to 120 producers, 72 processors, 72 consumers and 96 traders. The questionnaires were first pre- tested to validate the importance of the variables and the possible responses in addressing the objectives. 3.2.4. Data analysis Data collected were coded and analyzed using Statistical Package for Social Science (SPSS), 16th version. A combination of analysis that included percentages, descriptive statistics, and histograms were used to show the relationships between villages and districts in the study areas. To assess cassava production, processing and selection criteria, the Kendall’s coefficient of concordance (W) described by Mattson (1986) was used to rank the constraints. Coefficient of concordance (W) was computed as: 12 ∑ ?̅?2𝑖− 3𝑁(𝑁 + 1) 2 𝑊 = 𝑁(𝑁2 − 1) Where: W = Kendall’s value N = total sample size R = mean of the rank 20 University of Ghana http://ugspace.ug.edu.gh 3.4. RESULTS 3.4.1.Socio-economic characteristics of cassava farmers The socio-economic characteristics of cassava farmers are shown in Table 3.2. Three age groups were identified for each of the three districts with the adult group having the highest average percentage of 59.5% of the respondents. The gender structure of the respondents shows that the male had the highest percentage respondents with an average percentage of 93.2%.The educational levels of the respondents for each of the three districts shows that 34.7% of the respondents do not have any education level.The marital status of the respondents in the three districts shows that, the married group had the highest average percentage of 86.9%. Table 3.2:Percentage distributions of cassava farmers across the three study district Districts (%) Socioeconomic characteristics Bo Bombali Kenema Mean Age group Youth (18 – 35 years) 35.1 30.8 27.5 31.1 Adult (35 – 60 years) 62.2 56.4 60.0 59.5 Aged (above 60 years) 2.7 12.8 12.5 9.3 Gender Female 2.7 7. 7 10 .0 6. 8 Male 97.3 92.3 90.0 93.2 Educational level None 37.8 51 .3 15 .0 34 .7 Koranic 10.8 5.1 45.0 20.3 Primary 16.2 12.8 25.0 18.0 Junior Secondary School 24.3 23.1 10.0 19.1 Senior Secondary School 8.1 2.6 5.0 5.2 Tertiary 2.7 5.1 0.0 2.6 Marital status Single 16.2 12 .8 0. 0 9. 7 Married 81.1 79.5 100.0 86.9 Divorced 0.0 0.0 0.0 0.0 Widow / widower 0.0 2.6 0.0 0.9 21 University of Ghana http://ugspace.ug.edu.gh 3.4.2. Cassava production systems The study revealed that, in the Bo districts 89.19% of farmers grow only cassava while 10.81% grow and process cassava. In Bombali district, 51.28% grow only cassava while 48.72% grow and process. In Kenema district 82.50% of the farmers grow cassava only while 17.5% grows and process cassava (Figure 3.2). Grow and process Grow only 17.5 Kenema district 82.5 51.28 Bombali district 48.72 89.19 Bo district 10.81 Fig.3.2: Percentage distribution of type cassava growers in the three study districts 22 University of Ghana http://ugspace.ug.edu.gh 3.4.3. Farmer preference for starch content in the three study districts The percentage distribution of farmers’ preference for starch content in cassava across the three study districts is presented in Table 3.3. In the Bo district, 80.7% considered starch content in selecting cassava varieties for cultivation while 19.4% did not consider starch content in selecting cassava varieties for cultivation. In Bombali district, 79.5% considered starch content in selecting cassava varieties while 20.5% did not consider starch in selecting cassava varieties. In Kenema district, 67.6% considered starch in selecting cassava varieties and 24.1% did not consider starch content in selection of cassava varieties. A total of 80.7% of the respondents preferred high starch content in the cassava varieties used for cultivation in Bo district while 19.4% preferred moderate starch in the cassava varieties used for cultivation .In Bombali district, 74.4% preferred high starch content in the cassava varieties used for cultivation while in Kenema district, 97.3% preferred high starch content in the cassava varieties used for cultivation. In Bo district, 83.8% had cultivated cassava varieties with high starch content, 95% respondents in the Bombali district and 92.50% in Kenema district cultivated cassava varieties with high starch content. About 13.4% of the farmers in the three districts said they had easy access to cassava cutting of varieties with high starch content. Over 80% within the three districts (Bo, Kenema and Bombali) said they are willing to pay higher prices for cassava roots with high starch content. 23 University of Ghana http://ugspace.ug.edu.gh Table 3.3:Percentage distribution of farmer’s preference for starch content in cassava District (%) Farmer’s preferences Bo Bombali Kenema Mean 1.Consider starch content for selecting Yes 80.7 79.5 67.6 75.9 cassava varieties for cultivation No 19.4 20.5 32.4 24.1 2. Level of starch content preferred in High 80.7 74.4 97.3 84.1 cassava roots Moderate 19.4 25.6 2.7 15.9 Low 0.0 0.0 0.0 0.0 3.Grow cassava varieties with high starch Yes 83.8 100.0 92.5 92.1 content No 16.2 0.0 7.5 7.9 4.Have easy access to cuttings of cassava Yes 15.0 5.1 20.0 13.4 varieties with high starch content No 85.0 94.9 80.0 86.6 5.Willing to pay high price for cassava roots Yes 89.5 92.4 80.8 87.7 with high starch content No 10.5 7.6 19.2 12.3 3.4.4. Farmer preference for dry matter content in the three study districts Percentage distribution of farmers’preference for dry matter content in cassava across the three study districts revealed that, 80.7% of the respondents in the Bo district considered high dry matter content in cassava;79.9% in Bombali district while 67.6% in the Kenema district also considered high dry matter content in cassava. Determining the level of dry matter content in cassava roots shows that, in Bo district 80.7% of the respondents determined the level of dry matter content required in varieties grown, 74.4% in Bombali districts and 97.3% in Kenema district respectively. 24 University of Ghana http://ugspace.ug.edu.gh Farmers who grow cassava varieties with high dry matter content was, 83.8% of the respondents in Bo district, 95.0% in the Bombali district and 92.5% in Kenema District. However, 91.6% said they did not have easy access to cuttings of cassava varieties with high dry matter content in the three study area. In Bo, Bombali and Kenema districts, 95.5%, 96.4% and 96.6% of the respondent were willing to pay higher prices for cassava cuttings of varieties with high dry matter content (Table 3.4). Table 3.4: Percentage distribution of farmer’s preference for dry matter in cassava District (%) Farmer’s preferences Bo Bombali Kenema Mean 1.Consider dry matter content for selecting Yes 80.7 79.5 67.6 75.9 cassava varieties for cultivation No 19.4 20.5 32.4 24.1 High 80.7 74.4 97.3 84.1 2.Level of dry matter content preferred in Moderate 19.4 25.6 2.7 15.9 cassava roots Low 0.0 0.0 0.0 0.0 Yes 83.8 95.0 92.5 90.4 3.Grow cassava varieties with high dry matter content No 16.2 5.0 7.5 9.6 Yes 5.0 5.1 15.0 8.4 4.Have easy access to cuttings of cassava varieties with dry content No 95.0 94.9 85.0 91.6 Yes 95.5 96.4 90.8 94.2 5.Willing to pay high price for cassava roots with dry matter content No 4.5 3.6 9.2 15.8 25 University of Ghana http://ugspace.ug.edu.gh 3.4.5. Cassava processors preference for dry matter and starch content in cassava Analysis of cassava processors’ preferences for cassava dry matter and starch content in the three districts showed (Table 3.5) that, 94.74% of the processors processed cassava roots with high starch content in the Bo district, 95% in Bombali and 94% in Kenema districts. Ninety four percent of the respondents have easy access to cassava roots with high starch content in Bo district, 95% in Bombali district and 83.33% in Kenema district. A total of 83.3% respondents consider starch content in selecting cassava products for processing while in Bombali and Kenema districts, 90% and 83.33% respectively, were involed. In Bo district, 44.44% of the respondents prefer high starch content in the cassava varieties used for processing, 80% in the Bombali district and 58.3% in Kenema district. Processors also reported that, they have processed cassava varieties with high dry matter content 89.47% in Bo district, 95% in Bombali district and 91.57% in Kenema district. Ninety four percent in Bo district have easy access to cassava root with high dry matter content, 75% in Bombali district and 81.82% in Kenema district. In Bo district 94.12% consider dry matter content in selecting cassava for processing into products consumed, 92% in Bombali district and 90.9% in the Kenema district. In Bo district, 90.06% prefer high dry matter content in the cassava products they consumed, 80% in the Bombali district and 81.75% in Kenema district. 26 University of Ghana http://ugspace.ug.edu.gh Table 3.5:Percentage distribution of cassava processors for cassava roots with starch content and dry matter Cassava Bo Bombali Kenema traits (%) (%) (%) i. Have access to process cassava roots with high starch content 94.74 95.00 94.00 ii. Have easy access to cassava roots with high starch content 94.44 75.00 83.33 Starch iii. Consider starch content in selecting the cassava roots used for processing 83.33 90.00 83.33 iv. Prefer high starch content in the cassava roots used for processing 44.44 80.00 58.33 i. Have access to process cassava roots with high dry matter content 89.47 92.00 91.67 ii. Have easy access to cassava roots with high dry matter content 94.12 75.00 81.82 Dry matter iii. Consider dry matter content in selecting the cassava roots used for processing 94.12 90.00 90.91 iv. Prefer high dry matter content in the cassava roots used for processing 47.06 80.00 81.82 3.4.6. Cassava Trader’s preferences for starch and dry matter content in cassava products Table 3.6 shows percentage distribution of cassava traders’ preferences for starch and dry matter content. In the Bo district 88.89% of respondents sold cassava products with high starch content, 100% in Bombali district and 81.82% in Kenema. In the Bo district 87.50% of respondents have easy access to cassava products with high starch content, 100% in Bombali district and 81.25% in Kenema district. Traders who consider starch content in selecting cassava products sold represented 81.25% in Bo district, 95.83% in Bombali and 27 University of Ghana http://ugspace.ug.edu.gh 100% in Kenema district. In Bo district 43.73% of the respondents prefers high starch content in the cassava products sold, 29.17% in Bombali district and 22.22% in Kenema district. However, 61.11% of the traders in Bo district said they had sold cassava products with high dry matter content, 87.50% in the Bombali district and 95.45% in Kenema district. All traders in the Bo and Bombali districts have easy access to cassava products having high dry matter content with 95.24% in Kenema district. Hundred percent of respondents consider dry matter content in selecting cassava products for sale in Bo and Bombali districts and 95.24% in Kenema district. In Bo district, 81.82% prefer high dry matter content in the cassava products sold, 42.84% in Bombali district and 80.95% in Kenema district . Table 3.6: Percentage distribution of trader’s preferences for starch and dry matter in cassava products in the three study districts Cassava Bo Bombali Kenema traits (%) (%) (%) i. Have sold cassava products with high starch content 88.89 100.00 81.82 ii. Have easy access to cassava products with high starch content 87.50 100.00 100.00 Starch iii. Consider starch content in selecting the cassava products sold 81.25 95.83 100.00 iv. Prefer high starch content in the cassava products sold 43.75 29.17 22.22 i. Have sold cassava products with high dry matter content 61.11 87.50 95.45 ii. Have easy access to cassava products with high dry matter content 100.00 100.00 95.24 Dry matter iii. Consider dry matter content in selecting the cassava products sold 100.00 100.00 95.24 iv. Prefer high dry matter content in the cassava products sold 81.82 42.86 80.95 28 University of Ghana http://ugspace.ug.edu.gh 3.4.7. Consumer’s preference for cassava starch content and dry matter content The results for consumers preferences for cassava starch and dry matter content are shown in Table 3.7. In the Bo district 88.46% of the respondents had eaten cassava products with high starch content as compared to 100% in Bombali district and 84.38% in Kenema district. Hunderd percent had easy access to cassava products with high starch content in Bo district, 83.33% in Bombali district and 96.30% in Kenema district. In Bo district 86.96% of respondents considered starch content in selecting the cassava products while in Bombali and Kenema districts, 86.67% and 96.30% respectively,consider starch content. In Bo district, 52.17% of the respondents preferred high starch content in the cassava products eaten, 10% in Bombali district and 25.93% in the Kenema district. The study also showed that, 96.15% of the respondents in Bo district had eaten cassava products with high dry matter content, 100% in Bombali and Kenema districts. Ninety two percent had easy access to cassava products with high dry matter content in the Bo district, 86.67% in Bombali district and 96.88% in Kenema district. Hundred percent consider dry matter content in selecting cassava products eaten in Bo and Kenema districts and 80% in Bombali district. In Bo district,84% preferred high dry matter content in the cassava products they eat, 93.33% in Bombali district and 93.75% in Kenema district. 29 University of Ghana http://ugspace.ug.edu.gh Table 3.7: Percentage distribution of consumer’s preference for starch and dry matter incassava products across the three study districts Bo Bombali Kenema Cassava traits (%) (%) (%) i. Have eaten cassava products with high starch content 88.46 100.00 84.38 ii. Have easy access to cassava products with high starch 100.00 83.33 96.30 content Starch iii. Consider starch content in selecting the cassava products 86.96 86.67 96.30 eaten iv. Prefer high starch content in the cassava products eaten 52.17 10.00 25.93 i. Have eaten cassava products with high dry matter content 96.15 100.00 100.00 ii. Have easy access to cassava products with high dry matter 92.00 86.67 96.88 content Dry matter iii. Consider dry matter content in selecting the cassava products 100.00 80.00 100.00 eaten iv. Prefer high dry matter content in the cassava products eaten 84.00 93.33 93.75 3.4.8. Cassava products consumption by districts The study showed that gari, fufu, cassava roots and tho were highly consumed in the three study districts at 100%. On the other hand, cassava flour was only consumed in Kenema district with a high average percentage of 100%, while in Bo and Bombali districts consumption was 3.85% and 6.67 % respectively (Figure 3.3). 30 University of Ghana http://ugspace.ug.edu.gh 100 100 100 100 100 100 96.15 100 92.31 93.75 93.7590 90 84.38 80 73.08 70 60 50 40 30 20 6.67 10 3.85 0 Bo Bombali Kenema Fig. 3.3: Percentage distribution of respondents to Cassava products consumer across the three study district 3.4.9. Cassava production systems Percentage distribution of type of cassava growers and processors across the three study districts is shown in Figure 3.4. In the Bo district, 10.81% farmers cultivated only cassava, while 89.19% cultivated and processed. In Bombali district, 48.72% of respondents cultivated only cassava, 51.28% cultivated and processed cassava and in Kenema district, 82.50% farmers cultivated cassava while, 17.5% cultivated and processed. 31 Respondent (%) Cassava roots Gari Fufu HQCF Tho Cassava roots Gari Fufu HQCF Tho Cassava roots Gari Fufu HQCF Tho University of Ghana http://ugspace.ug.edu.gh Grow and process Buy and process Both 100 90 80 70 60 50 40 30 20 10 0 Bo Bomabli Kenema Fig. 3.4: Percentage distribution of type of cassava growers and processors across the three study districts. 3.4.10. Farmers’ selection criteria for cassava cultivars Table 3.8 shows cassava traits ranked by farmers across the three study districts. High yield was the most desirable trait farmers want to be incorporated into cassava breeding programmes followed by root size and root taste. However, in each district the ranking of desirable traits to be incorporated in cassava breeding programme varied. In Bo dstrict, high yield (1st), root size (2nd), dry matter content (3rd) and starch content (4th), were the most important traits. In Bombali districts, high yield (1st), root taste (2st), dry matter (3rd) and poundability (4th) were the most desired. In Kenema districts, high yield (1st), root taste (2st), maturity period (3rd), and poundability (4th), were the most desired. 32 Respondent (%) University of Ghana http://ugspace.ug.edu.gh Table 3.8: Cassava traits ranking by farmers across the three study districts Bo Bombali Kenema Overall Cassava traits Mean Rank Mean Rank Mean Rank Mean Rank High yield 2.35 1 3.74 1 2.28 1 2.79 1 Outer skin colour 5.86 7 6.23 8 7.68 8 6.59 8 Root size 3.38 2 4.67 4 4.4 5 4.15 2 Starch content 5.11 4 5 6 6.1 6 5.40 6 Dry matter content 4.86 3 4.33 3 6.5 7 5.23 5 Root taste 5.57 5 4.31 2 2.98 2 4.29 3 Resistance to pests 7.11 8 8.08 9 8.7 9 7.96 9 Resistance to disease 7.46 9 8.92 10 9.12 10 8.50 10 Poundability 7.73 10 4.67 4 4.05 4 5.48 7 Maturity period 5.57 5 5.05 7 3.2 3 4.61 4 P value <0.0001 <0.0001 <0.0001 <0.0001 Kendall’s W 0.321 0.322 0.655 0.433 Kendall’s W: Kendall’s coefficient of concordance 3.4.11. Processors’ selection criteria for cassava storage roots Cassava traits ranked by processors across the three study districts are shown in Table 3.9. Starch content was the most desirable trait processors want to be incorporated into cassava breeding programmes followed by dry matter content and root size. However, in each districts the ranking of desirable traits to be incorporated in cassava breeding programme varied: In Bo district, dry matter (1st), starch content (2nd), root size (3rd) and inner flesh colour (4st), were the most important. In Bombali district, starch content (1st), dry matter content (2nd), root size (3rd) and post physiological deterioration (4th) were the most desired. In the Kenema districts, starch content (1st), dry matter content (2nd), and root size (3rd) and ease of peeling (4th) were the most preferred. 33 University of Ghana http://ugspace.ug.edu.gh Table 3.9: Cassava traits ranking by processors across the study districts Bo Bombali Kenema Overall Cassava traits Mean Rank Mean Rank Mean Rank Mean Rank Root size 3.79 3 3.8 3 3.67 3 3.75 3 Starch content 3.05 2 1.8 1 3.25 1 2.70 1 Dry matter content 2.42 1 2.2 2 3.5 2 2.71 2 Ease of peeling 5.11 5 5.2 5 4 4 4.77 4 PPD 5.58 7 4.8 4 5.58 6 5.32 6 Fibre content 5.42 6 6.25 7 5.75 7 5.81 8 Age of tuber 6.26 8 6.45 8 4.5 5 5.74 7 Inner flesh colour 4.37 4 5.5 6 5.75 7 5.21 5 P value <0.0001 <0.0001 0.029 <0.0001 Kendall’s W 0.296 0.512 0.186 0.331 PPD: Post physiological deterioration, Kendall’s W: Kendall’s coefficient of concordance 3.4.12. Preferred varieties for different cassava products along the value chain The most preferred cassava varieties for different cassava products in the three study districts are shown in Table 3.10. SLICASS 4 and 3 MONTH were preferred for processing gari, SLICASS 4, ROCASS and 8 MONTH for starch processing, CARE, ROCASS and KABBAY for fufu, SLICASS 6 and WARIMA for cassava flour and CARE, WARIMA and BUTTER for processing tho. 34 University of Ghana http://ugspace.ug.edu.gh Table 3.10: Preferredcassava varieties for different cassava products in the three study districts Bo Bombali Kenema Product Variety % Variety % Variety % Gari SLICASS 4 64.7 3 MONTH 30.0 SLICASS 4 40.0 Fufu CARE 50.0 ROCASS 35.7 KABBAY 30.0 Starch SLICASS 4 100.0 ROCASS 50.0 8 MONTH 100 HQCF SLICASS 6 100.0 WARIMA 50.0 Tho CARE 66.7 WARIMA 20.0 BUTTER 50.0 3.4.13.SWOT analysis of cassava farmers The SWOT analysis for cassava farmers across the three study areas is presented in Table 3.11. Farmers strength ranged from 1.00% -100 %, weakness ranged from 1.00% - 97.50%, opportunities ranged from 2.50% - 94.87 % and threats ranged from 1.00% -95.00% . 35 University of Ghana http://ugspace.ug.edu.gh Table 3.11: SWOT analysis for cassava farmers in the three study districts Bo Bombali Kenema SWOT Analysis (%) (%) (%) Strength Agricultural land 97. 30 100 .00 100 .00 Improve planting materials 83.78 87.17 95.00 Family labour 62.16 97.44 90.00 Finance 35.13 1.00 15.00 Membership in FBO 18.92 2.56 1.00 Process cassava products 21.62 12.82 1.00 Weakness Lack of finance 94. 59 92. 31 85. 00 Lack of improve varieties 40.54 58.97 27.50 Lack of labour 35.14 20.51 22.50 Lack of training on improved agricultural practice 56.76 5.13 97.50 Lack of agro chemicals 64.86 94.87 67.50 Difficult access to land 8.11 2.56 1.00 Opportunities Development of improved cassava varieties 67. 57 89. 74 82. 50 Support from MAFFS 21.62 10.25 7.50 Availability of markets 70.27 94.87 90.00 Availability of processing centers 48.65 2.56 90.00 Microfinance 13.52 35.89 22.50 Support from NGOs 40.54 35.90 5.00 Strong link with research institutions 37.84 30.77 2.50 Threats Thieves 24. 33 17. 94 12. 50 Grasshoppers 83.79 89.75 92.50 Vertebrate pests 78.37 87.19 95.00 High transportation cost 81.08 48.72 92.50 Strong market competition 2.70 2.56 7.50 Fire outbreak 2.70 15.39 1.00 Land tenure 16.22 38.46 1.00 36 University of Ghana http://ugspace.ug.edu.gh 3.4.13. SWOT analysis of cassava processors The study showed that across the study areas, processors strength ranged from 1.00% - 95.00%, weakness ranged from 25.00% -100%, opportunities ranged from 10.00 % - 100% and threats ranged from 5.26% -94.74 % respectively.Table 3.12 37 University of Ghana http://ugspace.ug.edu.gh Table 3.12: SWOT analysis for cassava processorsin the three study districts Bo Bombali Kenema SWOT Analysis (%) (%) (%) Strengths Strong knowledge and experience in cassava processing 94. 74 95. 00 91. 67 Have access to labour for processing activities 94.74 75.00 91.67 Have access to credit and finance 26.32 1.00 8.33 Have access to Market and storage facilities for products 21.05 30.00 25.00 Strong linkages with farmers for tubers 26.32 80.00 50.00 Have access to processing equipment 36.84 20.00 33.33 Weakness Have limited access to market 52. 63 60. 00 66. 67 Have limited access finance and credit facilities 89.47 100.00 58.33 Use of local processing equipment 73.68 70.00 75.00 Lack of training on quality gari production 36.84 25.00 33.33 Inability to pay high transport fare for raw materials 47.37 45.00 66.67 Opportunities Availability of improved cassava varieties 89. 47 95. 00 100 .00 Strong linkages with VC actors 42.11 85.00 66.67 High demand for gari in local markets 84.21 85.00 75.00 Provision of training by NGO’s 15.79 10.00 25.00 Availability of processing centers with modern equipment facilities 10.53 25.00 33.33 Threat Increase in labour costs for gari activities 94.74 75.00 91.67 High cost for accessing improved processing equipment & inputs 84.21 65.00 91.67 Inadequate supply of raw materials 10.53 45.00 25.00 Theft 5.26 15.00 8.33 High interests rates on loans 47.37 40.00 16.67 Market diversity and competition with other food items 57.89 60.00 66.67 38 University of Ghana http://ugspace.ug.edu.gh 3.4 Discussion The findings of this study revealed stakeholders preference for starch content and dry matter content for different cassava varieties. The study also revealed that majority of the farmers were male and were dominated by married people in the three study districts.This could be due to the fact that majority of the rural people were challenged with the responsibility of early marriage to increase household labour size. Eighty nine percent of the respondents grow cassava only whilst 82.50% grow and process in the three study districts. This could be attributed to high level of cassava consumption in the districts as it serves as an important staple food crop which can be processed into various cassava-based products including boiled storage roots, gari, starch and cassava bread (very thin, small, and flat, round pieces) traded mainly in Sierra Leone (Latif et al., 2009). Farmers commonly select cassava varities with focus on high yield, root size, root taste, early maturity period and inner colour. The selection criteria reflects the importance of farmers’ needs and priorities, as well as the type of farming systems they practice. High yield, root taste and inner colour were selected because farmers believed that high yielding cultivars generate income. Similar results were reported by Ntumngia (2006) indicating that marketable roots,root size shape and colour of the skin, determine the demand and price for different cassava cultivars in the market. Farmers, however, preferred varieties that are resistance to pests and diseases as a way of ensuring food security for their households. Selection criteria were based on starch and dry matter contents. These selection criteria indicates that,cultivars with high dry matter and starch content are good in making high quality cassava based products (fufu, gari, tho and flour) and are more marketable.Cassava cultivars that are selected for the market, should, therefore, meet most of these qualities if farmers and processors have to stay competitive in the market and increase income from 39 University of Ghana http://ugspace.ug.edu.gh cassava. The combination of desired traits that meet their culinary, agronomic and other needs are based on local knowledge which is translated into their every day cultivar selection strategies and practices. Preference for the different cassava products (cassava roots, gari, fufu, and tho) in the study districts was high except for cassava flour that was low in the Bo and Bombali districts due to lack of market facilities in the districts. Several cassava cultivars were grown by farmers, with more than three cultivars identified in each of the surveyed districts. Each cultivar was selected for its special attributes preferred by processors. Strength, weakness, opportunities and threats faced by the farmers and processors were cross-cutting. These major cross-cutting issues identified by farmers as strength included agricultural land, improved planting materials and family labour.Weaknesses were lack of finance, lack of training on improved agricultural practices and lack of agro chemical. The opportunities were development and availability of improved cassava varieties and markets. The threats were grasshopper attack, high transportation cost and pest infestation. Strength for processors included knowledge and experience in cassava processing,easy access to labour for processing and strong linkages with the farmers for storage root.Their weaknesses include limited access to market, limited access to finance and credit facilities and use of local equipment. Opportunities were availability of improved cassava varieties, high demand for cassava products in local market and strong linkages with value chain actors. Threats identified were increase in labour cost for gari producing activities, high cost for improved processing equipment and inputs. 40 University of Ghana http://ugspace.ug.edu.gh 3.6. Conclusion Stakeholder’s preference for cassava dry matter content was identified. Farmers ranked high yielding cassava varieties as the highest priority in selecting cassava varieties followed by, root size, root taste and maturity period. Processors ranked starch content as the highest selection criteria followed by dry matter content, root size and ease of peeling. Several factors limiting cassava production and processing in the surveyed districts were also identified, key of which were diseases, especially cassava mosaic disease, grasshopper, lack of finance, and high transportation cost. Criteria for selecting cassava based products were also identified. To bridge the gap between stakeholders and breeders, participatory varietal selection and participatory plant breeding should be conducted to promote collaborations between stakeholders and breeders for developing new cassava varieties with high dry matter and starch contents. 41 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR 4.0. GENETIC DIVERSITY AND VARIABILITY STUDIES OF CASSAVA GENOTYPES FOR STARCH CONTENT, DRY MATTER CONTENT AND STORAGE ROOT YIELD 4.1. Introduction The starchy storage roots of cassava have become the most important source of dietary energy in sub -Saharan Africa (SSA) as they provide more returns per unit of input than any other staple crop (Fregene et al., 2000; Scott et al., 2000;Nassar, 2005). Cassava plant can grow and produce appreciable yields in poor soils with low fertility (Temegne et al., 2015a). According to Dixon et al. (2002), cassava genotypes respond differently to soil, climatic and biotic factors. Fukuda and Guevara (1998) reported that the evaluation of genetic diversity and variability in cassava is necessary and must be based on appropriate and recognized descriptors. Elias et al. (2001) also reported that morphological traits have a heritable genetic variation. Agro-morphological characterization had been used frequently in preliminary studies because they are fast and easy approach for assessing the extent of diversity among germplasm (Asare, 2011). Some of these morphological traits revealed the true diversity as perceived by farmers (Mckey et al., 2001, Pinton et al., 2001). Collard et al. (2005) reported that the use of molecular markers may permit the detection of genetic differences among closely related genotypes that are not affected by the environment. Characterization of accessions may, therefore, be more reliable if molecular markers are closely associated with morphological traits. SNPs markers are the most common used type of markers for genetic diversity and variability studies among species, because it represents difference in single nucleotide. This new sequencing and marker genotyping technology has 42 University of Ghana http://ugspace.ug.edu.gh accelerated the pace of genetic diversity research and gains in selection through molecular breeding. It is therefore, important to characterize and understand the genetic diversity and variability of cassava in order to select superior genotypes with desirable traits. The objectives of this study were to: 1. Assess the genetic diversity among cassava genotypes. 2. Assess variability for starch content, dry matter content and storage root yield 4.2 Materials and methods 4.2.1 Sources of Plant material One hundred and two diverse cassava genotypes (Table 4.1) which consisted of 82 white and 20 yellow accessions were evaluated to determine their genetic diversity out of a collection of 1110 genotypes maintained in the Sierra Leone germplasm bank. The plants were grown under field conditions in south of Sierra Leone from July 2015 to June 2016. 4.2.2 Experimental design and plot layout The trials were established at the experimental station of Njala Agricultural Research Centre (NARC) crops site Faya in the 2015 cropping season. Njala is situated at an elevation of 50 m above sea level, 8o06’N latitude and 12o06’W longitude. Stem cuttings of about 20 - 30 cm in length were planted in rows with 10 plants each per genotype at a spacing of 1 m x 1 m between and within rows using alpha lattice design with three replicates. No fertilizer or herbicide was applied. Hand weeding was done when ever necessary. 43 University of Ghana http://ugspace.ug.edu.gh 4.2.3 Agro-morphological characterization Based on morphological and agronomic descriptors of cassava by Fukuda et al. (2010), the quantitative and qualitative parameters were recorded at three, six, nine and twelve months after planting (MAP). A total of 22 morphological traits (11 quantitative and 11 qualitative) were collected at harvest (Table 4.2). 1. Harvest index (HI) was calculated as a ratio of fresh root yield to the total fresh biomass. 2. Reaction to Cassava Mosaic Disease (CMD) and Cassava Green Mite (CGM) were assessed at 3, 4 and 5 MAP using a scale of 1-5 as described by IITA (1990) to rate the genotypes for resistance to diseases and pests. 3. Starch extraction: Native starch extraction was carried out using a method described by Benesi (2005). 4. Percentage starch was calculated as: 𝐷𝑆𝑊 𝑆𝑡𝑎𝑟𝑐ℎ(%) = 𝑋100 𝐹𝑀 Where; FM= Fresh root weight; DSM= Dry starch weight 5. Root dry matter content (RDMC) determination: sampling for dry matter was done by selecting three representative roots from the bulk of roots harvested from 5 plants. Cassava roots were washed and shredded into pieces. A standard measure of 100 g weight of the fresh samples were taken and oven dried with forced draught oven. Samples were reweighed again to obtain a constant weight after 72 hrs at 65-70oC (Fukuda et al., 2010). 44 University of Ghana http://ugspace.ug.edu.gh Table 4.1: Descriptionof 102 cassava genotypes Genotype Storage root colour Genotype Storage root colour Genotype Storage root colour TR0991 white TR0188 white TR1523 white TR1827 white TR1782 white TR0218 white TR1036 white TR1069 white TR0043 white TR1013 white TR0075 white SLICASS6 white TR0038 white TR0575 white TR0142 white TR0768 white TR1696 white TR0813 white TR0519 white TR0127 white TR0971 white TR0523 white TR0329 white TR0593 white TR0092 white TR0864 white TR0589 white TR1198 yellow TR0302 white TR0345 white TR1028 yellow TR1005 white TR0135 white TR0949 yellow TR1162 white TR1436 white TR0300 yellow TR0159 white TR1326 yellow TR0356 white TR0006 white TR0454 white TR1769 white TR0255 white TR1736 white TR0024 white TR0435 white TR0724 white TR0275 white TR0428 white TR0310 white TR0694 yellow TR1791 white TR0417 white TR0912 white TR0288 white TR0120 white TR0105 white TR1716 yellow TR0164 white TR0171 white TR0432 white TR0263 white TR0488 white TR0297 yellow TR0590 white SLICASS4 white TR1167 white TR0382 yellow TR0746 white TR1175 white TR0453 white TR0740 white TR0821 white TR1154 white TR0204 white TR0657 white TR1097 white TR1518 yellow TR1035 yellow TR0545 white TR0064 white TR1618 yellow TR1041 yellow TR0455 white TR0613 white TR0745 white TR0779 white TR0358 white TR0017 white TR0820 white TR1796 yellow TR0591 white TR0915 white TR1288 white TR1159 white TR1824 white TR0667 white TR1776 white TR0812 white TR0256 white COCO white 45 University of Ghana http://ugspace.ug.edu.gh Table 4.2: Qualitative and quantitative traits used to characterize 102 cassava genotypes Qualitative Traits Quantitative Traits 1 Color of leaf vein (CLV) 1 Number of storage roots/plant (NSR) 2 Root taste (RT) 2 Harvest index (HI) 3 Cassava Mosaic disease 3 Storage root count 4 Color of root pulp (CRP) 4 Dry matter content (%) (DM 5 Lobe margins (LM) 5 Starch % (STC %) 6 Ease of peeling (EP) 6 Root yield per plant (RYPP) 7 Leaf colour (LC) 7 Harvesting Index (HI) 8 Colour of adult end branching (CEB) 8 Plant height (PH) 9 Root shape (RH) 9 Length of leaf lobe (LLL) 10 Shape of central leaflet (SCL) 10 Width of leaf lobe (WLL) 11 External color of storage root (ECSR) 11 Height at first Branching 4.2.4. Molecular characterization 4.2.4.1 DNA extraction DNA was extracted using the method proposed by Dellaporta et al.(1983) with a slight modification described by Rabbi et al. (2014). Freshly harvested apical leaves of about 200 mg of each accession were used. Grinding of the leaf samples was done in a 1.2 ml extraction tube using 400 µl extraction buffer and then placed in a 65 ºC water-bath for 25 min with gentle shaking. The tube was removed from the water bath and allowed to cool for 5-10 minutes. Proteins and polysaccharides were precipitated by adding 200 µl of ice-cold 5M potassium acetate and then mixed by gentle inversions (this was placed on ice for 20 min),350 µl chloroform: isomyl alcohol was added (24:1) and mixed gently with continuous rocking and centrifuge at 4000 g for 10 minutes. This was followed by the addition of RNase. The crude pellets were precipitated by transferring the upper layer to a new tube. One volume (400 µl) of ice-cold isopropanol was added and mixed gently for about 2-3 minutes and then 46 University of Ghana http://ugspace.ug.edu.gh chilled in -20 °C freezer for 10 minutes to enhance DNA precipitation. It was then centrifuged at 4500 g for 20 minutes and the supernatant was carefully discarded. 4.2.4.2 SNP genotyping Genotyping-by-sequencing analysis was determined as described by Elshire et al. (2011) and sequenced at the Institute of Genomic Diversity at Cornell University using the Illumina HiSeq 2500. The data were processed through cassava GBS production pipelines using TASSEL 5.0. A total of 8,600 SNP markers were used to assess the genetic diversity among the cassava accessions. Of these markers, resulting hap map files (SNPs) were filtered with minor allele frequency (MAF) of 0.001 and coverage of 10 xs. SNPs were further processed by removing those with MAF of less than 0.01 and loci with more than 40 % missing data SNP markers that were monomorphic or with more than 40% missing data were considered non- informative, and were removed from further analysis, as a result, only 5,600 informative SNP markers were retained for genetic diversity study. 4.2.5. Determination of physicochemical properties of cassava starches Chemical and functional characteristics were determined by using the AOAC (2006) method as describe below. Water absorption capacity (WAC) Hundred gram of starch was weighed, 15 ml distilled water was added in a 25 ml centrifuge tube. The tube was agitated on a vortex mixer for 2 minutes; the mixture was centrifuged at 400 rmp for 20 minutes. The supernatant was decanted and discarded by removing the adhering drops of water and reweighing the tube. Water absorption capacity was calculated by expressing water absorption capacity as the weight bound by 100 g dry starch. 47 University of Ghana http://ugspace.ug.edu.gh Determination of Amylose About 0.1 g of starch was weighed into centrifuge tubes containing 1 ml of 95% ethanol and 9 ml of NaOH. The tubes were capped and contents mixed thoroughly. The samples were heated for 10 minutes in a water bath to gelatinize the starch, and allowed to cool. Distilled water was added to make each tube up to 10 ml. This was left to stand for 20 mins to allow colour development and then vortexed and read at 620 nm. Percentage Amylose content was calculated as: % 𝐴𝑚𝑦𝑙𝑜𝑠𝑒 𝑜𝑓 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑥 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒 % 𝐴𝑚𝑦𝑙𝑜𝑠𝑒 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒 = 𝐴𝑏𝑠𝑜𝑟𝑏𝑎𝑛𝑐𝑒 𝑜𝑓 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 Determination of free sugar 0.020-0.025 g starch was weighed into centrifuge tubes. The starch powder was mixed with 1.0 ml of ethanol and 2.0 ml of distilled water was added. Also, 10 ml of hot ethanol was added and vortexed in a centrifuge for 10 mins at 2000 rpm.The supernatant solution was decanted into test tube and 7 ml of distilled to water was added to make it up to the 20 ml mark. Sugar content was determined as: (𝐴 − 𝐼) 𝑥 𝐷𝐹 𝑥 𝑉 % 𝑆𝑢𝑔𝑎𝑟 = 𝑥 100 𝐵 𝑥 𝑊 𝑥 106 Where A = Absorbance of sample I = Intercept of sample D.F = Dilution factor (depends on aliquot taken for assay) V = Volume B = Slope of the standard curve W = Weight of the sample Determination of starch content The residue from sugar analysis was used by adding 7.5 ml of perchloric and then allowed to stand for 1 hour. The hydrolysate was diluted with 17.5 ml distilled water to make up to 25 48 University of Ghana http://ugspace.ug.edu.gh ml. The content was filtered and 0.05 ml of the filtrate was diluted with 0.95 ml of distilled water and vortexed for assay. The starch content was determined as: (𝐴 − 𝐼) 𝑥 𝐷𝐹 𝑥 𝑉 𝑥 0.9 𝑥 100 % 𝑆𝑡𝑎𝑟𝑐ℎ = 𝐵 𝑥 𝑊 𝑥 106 Where A = Absorbance of sample I = Intercept of sample D.F = Dilution factor (depends on aliquot taken for assay) V = Volume B = Slope of the standard curve W = Weight of the sample 4.3 Data analysis Qualitative and quantitative measurements The morphological data were analysed using SAS 9.4 software version. Data on qualitative and quantitative traits were analysed separately. Analysis of variance (ANOVA) was used to assess the differences between cultivars. The percentage distribution of the different variants of qualitative traits was determined, PCA was used to explore the links among the qualitative traits. Quantitative variables were analyzed using means,coefficient of variation. Principal component analysis (PCA) and correlation matrices were used to determine the relationships among the traits. The organization and structure of the morphological variability were visualized using the Ascending Hierarchical Clustering (AHC) to plot a dendrogram. 49 University of Ghana http://ugspace.ug.edu.gh Molecular data The genetic analysis package Power Marker version 3.0 (Liu and Muse, 2005) was used to generate the following statistics: number of alleles per locus, major allele frequency, expected heterozygosity (HE), polymorphic information content (PIC) and gene diversity (GD) (Bostein and White, 1980). PIC values were calculated with the equation: PIC=1-ΣP2i－Σ 2P2i P2J. Identification of the cassava varieties was performed using two methods (i) complementary clustering approaches: (ii) pairwise distance-based hierarchical clustering. Physiochemical data on cassava starch were analysed using SAS 9.4. Means were calculated for each of the traits and the least significant difference (LSD) was used to test for the difference within and among the genotypes at significance difference of 5%. Correlation coefficient was calculated to determine the relationships among different starch traits. 4.4. Results 4.4.1. Genetic diversity using morphological characterization 4.4.1.1. Frequency distribution of accessions according to qualitative traits Frequency distributions of the qualitative traits are presented in Fig. 4.1 and Fig 4.2 Genetic variability was observed among the 102 cassava accessions for all of the variables evaluated. The results showed that 79.4% of the accessions had sweet root taste, 14.7% were classified as intermediate and 5.9% had bitter root. Approximately 90.1% of accessions were easy to peel while 9.8% were difficult to peel. In terms of colour of end branching, 84.3%, 12.7% and 2.9% were green, green purple and purple respectively. More than 50% of the accessions had green leaf vein, 27% had reddish-green and 6% had reddish-green in less than half of the lobe. Lobe margin of the 52% of the accessions was smooth while 48% had 50 University of Ghana http://ugspace.ug.edu.gh winding. About (70.6%) of the accession had white root pulp, 19.6% had cream root pulp and 9.8% had yellow root pulp. Over 90% of the accessions had green leaves, 48% light green and 44% dark green and 93% had sweet roots. The main shapes of the central leaflet among the accessions were 29.4% pandurate, 27.5% linear pyramidal and 24.5% oblong- lanceocate. 51 University of Ghana http://ugspace.ug.edu.gh 100 100 80 80 80 60 60 60 40 40 40 20 20 0 20 Easy Difficult 0 0 Sweet Intermediate Bitter Cream White Yellow Ease of peeling Colour of root pup Root taste 100.0 70 100 80.0 60 80 50 60.0 40 60 30 40.0 20 40 10 20.0 0 20 0.0 0 colour of end branching colour of leaf vein Root shape Fig. 4.1: Percentage distribution of six cassava qualitative traits 52 University of Ghana http://ugspace.ug.edu.gh 35 80 30 70 25 60 50 53 20 40 52 15 30 10 20 51 5 10 50 0 0 49 48 47 46 Winding Smooth Shape of central leaflet CMD severity month after planting Lobe margine 80 60 70 50 60 50 40 40 30 30 20 20 10 10 0 Light Dark brown White or 0 brown cream Light green Dark green Purple External colour of storage root green Leaf colour Fig. 4.2 : Percentage distribution of five cassava qualitative traits 53 Linear-pandurate Linear-piramidal Pandurate Oblong-lanceolate Straight or linear Lanceolate Obovate-lanceolate Ovoid University of Ghana http://ugspace.ug.edu.gh 4.4.1.2. Analysis of variance for eleven qualitative traits Table 4.3 shows the probability values and means of the analysis of variance.There was significant difference among the 102 genotypes for all the traits except for CMD where the accessions did not differ. CV ranged from 2.8 % for shape of central leaflet to 51.8% for CMD. 54 University of Ghana http://ugspace.ug.edu.gh Table 4.3: Probability values, means and coefficient of variation of qualitative traits of 102 cassava genotypes Source RT ECSR CRP ES CLV CMD RS LM LC CEB SCL Genotype <.0001 <.0001 <.0001 <.0001 <.0001 0.2298 <.0001 <.0001 <.0001 <.0001 <.0001 Mean 1.24 3.6 1.4 1.1 3.8 1.24 2.3 4.9 3.4 3.07 6.15 CV % 6.5 3.15 12.2 10.3 3.0 51.8 4.9 6.6 3.38 5.34 2.8 Significant at alpha = 0.05. RT = Root taste , ECSR = External colour of storage root , CRP = Color of root pulp ,ES = Ease of peeling, CLV = Color of leaf vein, CMD = Cassava mosaic disease ,RS = Root shape , LM = Lobe margins, LC = Leaf colour , CEB = Color of end branches of adult plant , SCL = Shape of central leaflet (SCL) 55 University of Ghana http://ugspace.ug.edu.gh 4.4.1.3. Principal component analysis of qualitative characters The eigenvalues and percentage variations of the principal component analysis are presented in Tables 4.4. Seven principal components that accounted for 79.03% of the total variation among the genotypes were identified. The first PC axis with eigenvalue of 1.73 accounted for 15.76 % of the total variation where the second, third and the forth PC axis with eigenvalues of 1.70, 1.35 and 1.14 accounted for 15.43%, 12.24% and 10.38% of the total variation, respectively. The fifth, sixth and seventh PC axis with eigenvalues of 0.99, 0.97 and 0.83 accounted for 8.99%, 8.84% and 7.43% of the total variation ,respectively. The first principal component with reference to its high factor loadings was positively associated with traits such as root taste, colour of root pulp, ease of peeling, and root shape. The second PC was associated with leaf and storage root characteristics (Root taste, leaf colour, and colour of end branching); the third PC was associated with external colour of storage root, ease of peeling, colour of leaf vein and shape of central leaf lobe while the forth PC was associated with traits related to storage root characteristics (colour of root pulp, external colour of storage root, and root shape) colour of leaf vein and cassava mosaic disease. The fifth PC was associated with characteristics such as root taste, cassava mosaic disease, root shape, lobe margin and colour of end branching, the sixth PC was also associated with external colour of storage root, cassava mosaic disease and lobe margin and the seventh PC was also associated with storage (root colour of root pulp and root shape) and Cassava mosaic disease. 56 University of Ghana http://ugspace.ug.edu.gh Table 4.4: Principal component analysis, eigenvalues and percentage variation of eleven qualitative traits of 102 cassava genotypes Traits Prin1 Prin2 Prin3 Prin4 Prin5 Prin6 Prin7 RT 0.43 -0.37 -0.04 0.22 0.32 0.09 -0.25 ECSR -0.14 0.26 -0.32 0.52 -0.05 0.41 0.02 CRP 0.48 0.10 0.12 -0.41 0.06 -0.02 0.48 ES 0.47 -0.22 0.33 0.29 0.12 0.15 -0.18 CLV -0.17 0.00 0.51 0.36 -0.27 0.02 0.21 CMD -0.18 0.13 0.19 0.33 0.67 -0.31 0.44 RS -0.41 -0.12 0.21 -0.30 0.36 -0.14 -0.46 LM -0.21 -0.18 0.01 -0.28 0.30 0.78 0.27 LC 0.22 0.60 0.03 -0.12 0.05 0.01 -0.07 CEB 0.12 0.50 -0.07 0.03 0.33 0.13 -0.32 SCL 0.03 -0.23 -0.65 0.08 0.15 -0.20 0.19 Eigenvalue 1.73 1.70 1.35 1.14 0.99 0.97 0.82 Proportion of variance (%) 15.76 15.43 12.24 10.38 8.99 8.84 7.43 Cumulative variance (%) 15.76 31.19 43.43 53.81 62.80 71.64 79.07 Valves in bold represent significant traits in the various principal components RT = Root taste , ECSR = External colour of storage root , CRP = Color of root pulp ,ES = Ease of peeling, CLV = Color of leaf vein, CMD = Cassava mosaic disease ,RS = Root shape , LM = Lobe margins, LC = Leaf colour , CEB = Color of end branches of adult plant , SCL = Shape of central leaflet (SCL) 57 University of Ghana http://ugspace.ug.edu.gh 4.4.1.4.Means values and correlation coefficients for the eleven quantitative traits The mean values for harvest index, average yield per plant, dry matter content, number of storage root and starch percentage were 0.5, 1.6 kg, 30.9%, 59.8 and 23.9% respectively. Significant differences were observed among the 102 genotypes for all the traits (Table 4.5). Significant and positive correlation were observed between average yield per plant and harvest index (r = 0.76), length of leaf and petiole length, number of storage roots per plant and average yield per plant (r = 0.58), starch percentage and dry matter content (r = 0.99), as well as height at first branching and plant height (r = 0.45) Table 4.6. 58 University of Ghana http://ugspace.ug.edu.gh Table 4.5: Probability values, means and coefficient of variation of quantitative traits of 102 cassava genotypes Source HI AY PL LL WLB DM NSR SCT PH NLL HFB Genotype <.0001 <.0001 <.0001 <.0001 0.9709 0.214 <.0001 0.3192 0.7652 <.0001 0.0098 Mean 0.45 1.56 13.5 10.6 4.4 30.9 59.8 23.9 149.1 5.8 58.9 CV 19.1 32.7 20.9 11.6 100.4 8.8 23.1 11.8 17.55 17.3 55.1 Significant at alpha = 0.05. HI = Harvest index, AY = Average yield per plant (kg) ,PL = Petiole length (cm), LL = Length of leaf lobe (cm), WLB = Width of leaf lobe (cm), DM = Dry matter %, NSR = Number of storage roots (count), SCT = Starch percentage, PH = Plant height (cm), HFB = Height at first branching (cm),NLL = number of number of leaf lobe 59 University of Ghana http://ugspace.ug.edu.gh Table 4.6: Correlation coefficients among 11 quantitative traits of 102 cassva genotypes HI AYPP PL LII WLFB DM NSR SCT PH HFB NLL HI 1.00 AYPP 0.76 1.0 0 PL 0.01 0.11 1.0 0 LII 0.14 0.23 0.38 1.0 0 WLB -0.19 -0.01 0.23 0.36 1.0 0 DM 0.08 0.00 -0.04 0.01 -0.16 1.0 0 NSR 0.33 0.58 0.14 0.19 0.01 0.03 1.0 0 STC% 0.08 0.00 -0.05 0.00 -0.16 0.99 0.05 1.0 0 PH 0.16 0.03 0.06 -0.18 -0.17 0.00 -0.03 -0.03 1.0 0 HFB 0.26 0.24 -0.04 -0.27 -0.21 0.06 0.08 0.05 0.45 1.0 0 NLL 0.05 -0.03 0.31 0.43 0.19 0.30 0.11 0.28 0.07 -0.03 1.0 0 Significant at alpha = 0.05. HI = Harvest index, AYPP = Average yield per plant (kg) ,PL = Petiole length (cm), LL = Length of leaf lobe (cm), WLB = Width of leaf lobe (cm), DM = Dry matter %, NSR = Number of storage roots (count), SCT = Starch percentage, PH = Plant height (cm), HFB = Height at first branching (cm),NLL = number of number of leaf lobe (c 60 University of Ghana http://ugspace.ug.edu.gh 4.4.1.5. Representation of variables of quantitative traits The result revealed that the four main components accounted for 72.30% of the total variation among the genotypes. The first factorial plane contains 22.18% of the variance. The variables that significantly correlated with axis 1 are: harvest index (47%), average yield per plant (49%), dry matter content (30%), number of storage roots (40%), and starch percentage (30.0%). The variables that were significantly correlated with axis 2 are: petiole length (37%), length of leaf lobe (51%), width of leaf lobe (47%) and height at first branching (- 37%). The variables significantly related to axis 3 are: average yield per plant (-35%), dry matter content (55%), starch content percentage (55%) and number of leaf lobes (30%). The variables significantly correlated with axis 4 are: petiole length (40%), plant height (64%), height at first branching (40%) and number of leaf lobe (36%) Table 4.7. 61 University of Ghana http://ugspace.ug.edu.gh Table 4.7: Principal component analysis, eigenvalues and percentage variation of eleven quantitative traits of 102 cassava genotypes Traits Prin1 Prin2 Prin3 Prin4 HI 0.47 -0.11 -0.28 -0.13 AYPP 0.49 0.02 -0.35 -0.23 PL 0.17 0.37 0.00 0.40 LII 0.24 0.51 0.06 0.02 WLFB -0.05 0.47 0.03 0.10 DM 0.30 -0.22 0.55 -0.07 NSR 0.40 0.10 -0.20 -0.21 STC 0.30 -0.21 0.55 -0.10 PH 0.10 -0.27 -0.14 0.64 HFB 0.19 -0.37 -0.20 0.40 NLL 0.26 0.26 0.30 0.36 Eigenvalue 2.44 2.14 2.07 1.31 Proportion of variance (%) 22.18 19.42 18.83 11.87 Cumulative Variance (%) 22.18 41.60 60.43 72.30 Figures in bold represent significant traits in the various principal components HI = Harvest index, AYPP = Average yield per plant (kg), PL = Petiole length (cm), LL = Length of leaf lobe (cm), WLB = Width of leaf lobe (cm), DM = Dry matter %, NSR = Number of storage roots (count), STC = Starch (%), PH = Plant height (cm), HFB = Height at first branching (cm), NLL = Number of leaf lobe (count) 62 University of Ghana http://ugspace.ug.edu.gh 4.4.2. Hierarchical clustering analysis 4.4.2.1.Dendrogram of qualitative traits The genetic similarity for the eleven qualitative traits ranged from zero to one with a mean similarity of 0.10. The cassava genotypes were grouped into five distinct clusters at 0.06 similarities. Groups III, IV and V have a high number of genotypes with 55, 22 and 12, respectively. Ten and 3 individuals were in clusters II and I (Figure 4.2 and Table 4.8). 63 University of Ghana http://ugspace.ug.edu.gh I II III IV V Fig. 4.4: Ward method of classification of genotypes based on eleven qualitative traits 64 University of Ghana http://ugspace.ug.edu.gh Table 4.8: Grouping of 102 genotypes based on qualitative traits CLUSTER GENOTYPES CLUSTER SIZE CLUSTER 1 TR1523, TR0300, TR0218 3 CLUSTER 2 TR1069, TR0949, TR1782, TR1013, TR0428, TR0523, TR1736, TR0255, TR1036, TR0038, 10 CLUSTER 3 TR0188, TR1159, TR0256, TR0135, TR1167, TR0915, TR0813, TR0667, TR0745, TR0120, TR0694, 55 TR0382, TR0991, TR0453, TR0288, TR0164, TR0417, TR0105, TR0590, TR1288, TR0589, TR0545, TR1796, TR0092, TR0971, TR0820, TR0657, TR1791, TR0746, TR0358, TR1618, TR0310, TR0159, TR0740, TR1776, TR1518, TR0064, TR0779, TR0488, TR1827, TR0127, TR0593, TR1436, TR0435, TR1097, TR1154, TR0912, TR1824, TR0006, TR0240, TR0545 TR0127, TR0575, TR0263, TR0768 CLUSTER 4 TR1198, TR1696, TR0519, TR1769, TR0356, TR1035 TR0821, TR0454, TR0297, TR0171, TR0075, 22 TR0024, TR1326, TR0345, TR0142, TR0864, TR0591, TR1041, TR0017, SLICASS6, TR0043, SLICASS4 CLUSTER 5 TR1162, TR1716, TR0432, TR1005, TR0329, TR0724, TR0302, TR1028, TR1157, TR0812, TR0275, 12 COCO 65 University of Ghana http://ugspace.ug.edu.gh 4.4.2.2. Dendrogram of quantitative traits Hierarchical classification of quantitative traits grouped genotypes into four classes almost with the same characteristics as a function of the variables (Figure 4.3, Table 4.9). The genetic similarity for the eleven quantitative traits ranged from zero to one with a mean similarity of 0.10. Cluster I contains 19 genotypes, cluster II 12 genotypes, cluster III 46 genotypes and cluster IV 25 genotypes. 66 University of Ghana http://ugspace.ug.edu.gh I II III IV Fig. 4.5: Ward method of classification of genotypes based on eleven quantitative 67 University of Ghana http://ugspace.ug.edu.gh Table 4.9: Grouping for 102 genotypes based on quantitative traits CLUSTERS GENOTYPES CLUSTER SIZE CLUSTER 1 TR1796, TR1791, TR1097, TR1326, TR0740, TR1782, TR0435, TR0263, TR0356, TR0300, TR0545, TR1036, 19 TR0204, TR1013, TR0302, TR1028, TR0275, TR0694, TR0142 CLUSTER 2 T R1035, TR0745, TR0667, TR0523, TR0488, TR0288, TR0171, TR1776, TR1523, TR0590, TR0453, TR0024, 12 CLUSTER 3 TR1736, TR1198, TR0768, TR0075, TR1162, TR1824, TR0820, TR0428, TR0971, TR0813, TR1159, TR0218, 46 TR0043, TR0991, TR0949, TR0135, TR1154, TR0915, TR0038, TR1518, TR0912, TR0345, TR0821, TR0589, TR0164, TR0779, TR0017, TR0006, TR0519, TR1069, TR0575, TR0382, TR0382, TR0432, TR0256, TR1618, TR1167, TR0864, TR0746, TR1696, TR1157, TR0454, TR0105, TR1716, TR0064, SLICASS4 CLUSTER 4 TR1769, TR0358, TR1436, TR0591, TR0188, TR0657, TR0092, TR0329, TR1288, TR0593, TR0724, 25 SLICASS6, TR0455, TR0255, TR0812, TR0159, TR0297, TR0310, TR0127, TR1827, TR0120, TR1005, TR0417, TR1041, COCO. 68 University of Ghana http://ugspace.ug.edu.gh 4.4.2.3. Genetic diversity analysis using SNP markers Summary statistics for major allele frequency, number of alleles observed, gene diversity,expected heterozygosity and polymorphic information content are presented in (Table 4.10). A total of 1.000 alleles with an average of 1 allele per locus were observed. The observed expected heterozygosity per individual SNP locus ranged from 0.26 at the CARK7ANXX:6:529049, CARK7ANXX:6:529107 and CARK7ANXX:6:529118 locus to 0.46 at the CARK7ANXX:6:529076 and CARK7ANXX:6:529116 locus. The major allele frequency of all the markers used was generally below 0.95, indicating that they were all polymorphic. The gene diversity ranged from 0.13 at the CARK7ANXX:6:529118 locus to 0.23 at the CARK7ANXX:6:529076 and CARK7ANXX:6:529116 locus. PIC values ranged from 0.10 at the CARK7ANXX:6:529049, CARK7ANXX:6:529068, CARK7ANXX:6:529034, CARK7ANXX:6:529017 and CARK7ANXX:6:529118 locus to 0.17 at CARK7ANXX: 6:529076 locus. 69 University of Ghana http://ugspace.ug.edu.gh Table 4.10: Summary statistics of genetic variation using 5,600 SNP markers among 102 cassava accessions SNP Marker MAF Na GD Ho PIC SRL3K_10:CARK7ANXX:6:529038 0.83 1.35 0.17 0.35 0.13 SRL3K_11:CARK7ANXX:6:529039 0.82 1.37 0.19 0.37 0.14 SRL3K_12:CARK7ANXX:6:529040 0.82 1.35 0.18 0.35 0.13 SRL3K_13:CARK7ANXX:6:529041 0.80 1.40 0.20 0.40 0.15 SRL3K_14:CARK7ANXX:6:529042 0.82 1.36 0.18 0.37 0.14 SRL3K_15:CARK7ANXX:6:529043 0.82 1.36 0.18 0.36 0.13 SRL3K_16:CARK7ANXX:6:529044 0.82 1.37 0.18 0.37 0.14 SRL3K_17:CARK7ANXX:6:529045 0.82 1.35 0.18 0.35 0.13 SRL3K_18:CARK7ANXX:6:529046 0.82 1.35 0.18 0.35 0.13 SRL3K_19:CARK7ANXX:6:529047 0.83 1.35 0.18 0.35 0.13 SRL3K_2:CARK7ANXX:6:529030 0.86 1.29 0.14 0.29 0.11 SRL3K_20:CARK7ANXX:6:529048 0.81 1.39 0.19 0.39 0.15 SRL3K_21:CARK7ANXX:6:529049 0.87 1.26 0.13 0.26 0.10 SRL3K_22:CARK7ANXX:6:529050 0.83 1.34 0.17 0.34 0.13 SRL3K_23:CARK7ANXX:6:529051 0.86 1.29 0.14 0.29 0.11 SRL3K_24:CARK7ANXX:6:529052 0.81 1.39 0.19 0.39 0.15 SRL3K_25:CARK7ANXX:6:529053 0.82 1.37 0.18 0.37 0.14 SRL3K_26:CARK7ANXX:6:529054 0.79 1.42 0.20 0.42 0.16 SRL3K_27:CARK7ANXX:6:529055 0.82 1.37 0.18 0.37 0.14 SRL3K_28:CARK7ANXX:6:529056 0.83 1.34 0.17 0.34 0.13 SRL3K_29:CARK7ANXX:6:529057 0.82 1.37 0.18 0.37 0.14 SRL3K_3:CARK7ANXX:6:529031 0.81 1.38 0.19 0.38 0.14 SRL3K_30:CARK7ANXX:6:529058 0.82 1.36 0.18 0.36 0.14 SRL3K_31:CARK7ANXX:6:529059 0.84 1.32 0.16 0.32 0.12 SRL3K_32:CARK7ANXX:6:529060 0.82 1.36 0.18 0.36 0.13 SRL3K_33:CARK7ANXX:6:529061 0.79 1.42 0.21 0.42 0.16 SRL3K_34:CARK7ANXX:6:529062 0.82 1.37 0.18 0.37 0.14 SRL3K_35:CARK7ANXX:6:529063 0.82 1.37 0.19 0.37 0.15 SRL3K_36:CARK7ANXX:6:529064 0.80 1.39 0.20 0.39 0.15 SRL3K_37:CARK7ANXX:6:529065 0.83 1.35 0.18 0.35 0.13 SRL3K_38:CARK7ANXX:6:529066 0.82 1.36 0.18 0.36 0.14 SRL3K_39:CARK7ANXX:6:529067 0.81 1.37 0.19 0.37 0.14 SRL3K_4:CARK7ANXX:6:529032 0.83 1.34 0.17 0.34 0.13 SRL3K_40:CARK7ANXX:6:529068 0.86 1.27 0.14 0.27 0.10 70 University of Ghana http://ugspace.ug.edu.gh Table 4.10: (cont’d) Summary statistics of genetic variation using 5,600 SNP markers among 96 cassava accessions SNP Marker MAF Na GD He PIC SRL3K_41:CARK7ANXX:6:529069 0.82 1.37 0.18 0.37 0.14 SRL3K_42:CARK7ANXX:6:529070 0.80 1.40 0.20 0.40 0.15 SRL3K_43:CARK7ANXX:6:529071 0.82 1.36 0.18 0.36 0.13 SRL3K_44:CARK7ANXX:6:529072 0.83 1.35 0.17 0.35 0.13 SRL3K_45:CARK7ANXX:6:529073 0.82 1.36 0.18 0.36 0.13 SRL3K_46:CARK7ANXX:6:529074 0.78 1.44 0.22 0.44 0.16 SRL3K_47:CARK7ANXX:6:529075 0.82 1.36 0.18 0.36 0.14 SRL3K_48:CARK7ANXX:6:529076 0.77 1.46 0.23 0.46 0.17 SRL3K_49:CARK7ANXX:6:529077 0.81 1.38 0.19 0.38 0.14 SRL3K_5:CARK7ANXX:6:529033 0.80 1.40 0.20 0.40 0.15 SRL3K_50:CARK7ANXX:6:529078 0.83 1.35 0.17 0.35 0.13 SRL3K_51:CARK7ANXX:6:529079 0.83 1.35 0.17 0.35 0.13 SRL3K_52:CARK7ANXX:6:529080 0.83 1.34 0.17 0.34 0.13 SRL3K_53:CARK7ANXX:6:529081 0.81 1.38 0.19 0.38 0.14 SRL3K_54:CARK7ANXX:6:529082 0.82 1.35 0.18 0.35 0.13 SRL3K_55:CARK7ANXX:6:529083 0.82 1.36 0.18 0.36 0.13 SRL3K_56:CARK7ANXX:6:529084 0.82 1.35 0.18 0.35 0.13 SRL3K_57:CARK7ANXX:6:529085 0.81 1.36 0.18 0.36 0.14 SRL3K_58:CARK7ANXX:6:529086 0.82 1.36 0.18 0.36 0.13 SRL3K_59:CARK7ANXX:6:529087 0.82 1.35 0.18 0.35 0.13 SRL3K_6:CARK7ANXX:6:529034 0.87 1.27 0.13 0.27 0.10 SRL3K_60:CARK7ANXX:6:529088 0.82 1.35 0.18 0.35 0.13 SRL3K_61:CARK7ANXX:6:529089 0.82 1.36 0.18 0.36 0.14 SRL3K_62:CARK7ANXX:6:529090 0.82 1.37 0.19 0.37 0.14 SRL3K_63:CARK7ANXX:6:529091 0.83 1.35 0.17 0.35 0.13 SRL3K_64:CARK7ANXX:6:529092 0.82 1.36 0.18 0.36 0.14 SRL3K_65:CARK7ANXX:6:529093 0.83 1.34 0.17 0.34 0.13 SRL3K_66:CARK7ANXX:6:529094 0.81 1.38 0.19 0.38 0.14 SRL3K_67:CARK7ANXX:6:529095 0.82 1.36 0.18 0.36 0.14 SRL3K_68:CARK7ANXX:6:529096 0.82 1.35 0.18 0.35 0.13 SRL3K_69:CARK7ANXX:6:529097 0.82 1.37 0.18 0.37 0.14 SRL3K_7:CARK7ANXX:6:529035 0.80 1.40 0.20 0.40 0.15 SRL3K_70:CARK7ANXX:6:529098 0.82 1.36 0.18 0.36 0.13 SRL3K_71:CARK7ANXX:6:529099 0.82 1.35 0.18 0.35 0.13 SRL3K_72:CARK7ANXX:6:529100 0.82 1.35 0.18 0.35 0.13 SRL3K_73:CARK7ANXX:6:529101 0.82 1.35 0.18 0.35 0.13 SRL3K_74:CARK7ANXX:6:529102 0.82 1.36 0.18 0.36 0.14 71 University of Ghana http://ugspace.ug.edu.gh Table 4.10: (cont’d) Summary statistics of genetic variation using 5,600 SNP markers among 96 cassava accessions SNP Marker MAF Na GD He PIC SRL3K_75:CARK7ANXX:6:529103 0.81 1.38 0.19 0.38 0.14 SRL3K_76:CARK7ANXX:6:529104 0.86 1.28 0.14 0.28 0.11 SRL3K_77:CARK7ANXX:6:529105 0.83 1.35 0.18 0.35 0.13 SRL3K_78:CARK7ANXX:6:529106 0.82 1.36 0.18 0.36 0.14 SRL3K_79:CARK7ANXX:6:529107 0.87 1.26 0.13 0.26 0.10 SRL3K_8:CARK7ANXX:6:529036 0.82 1.36 0.18 0.36 0.14 SRL3K_80:CARK7ANXX:6:529108 0.82 1.36 0.18 0.36 0.13 SRL3K_81:CARK7ANXX:6:529109 0.85 1.31 0.15 0.31 0.12 SRL3K_82:CARK7ANXX:6:529110 0.83 1.35 0.17 0.35 0.13 SRL3K_83:CARK7ANXX:6:529111 0.82 1.37 0.18 0.37 0.14 SRL3K_84:CARK7ANXX:6:529112 0.82 1.35 0.18 0.35 0.13 SRL3K_85:CARK7ANXX:6:529113 0.82 1.35 0.18 0.35 0.13 SRL3K_86:CARK7ANXX:6:529114 0.83 1.35 0.17 0.35 0.13 SRL3K_87:CARK7ANXX:6:529115 0.83 1.34 0.17 0.34 0.13 SRL3K_88:CARK7ANXX:6:529116 0.77 1.46 0.23 0.46 0.17 SRL3K_89:CARK7ANXX:6:529117 0.82 1.35 0.18 0.35 0.13 SRL3K_9:CARK7ANXX:6:529037 0.82 1.35 0.18 0.36 0.13 SRL3K_90:CARK7ANXX:6:529118 0.87 1.26 0.13 0.26 0.10 SRL3K_91:CARK7ANXX:6:529119 0.82 1.35 0.18 0.35 0.13 SRL3K_92:CARK7ANXX:6:529029 0.83 1.35 0.17 0.35 0.13 SRL3K_93:CARK7ANXX:6:529121 0.84 1.32 0.16 0.32 0.12 SRL3K_94:CARK7ANXX:6:529122 0.82 1.36 0.18 0.36 0.13 SRL3K_95:CARK7ANXX:6:529123 0.82 1.37 0.18 0.37 0.14 SRL3K_96:CARK7ANXX:6:529124 0.83 1.33 0.17 0.33 0.13 MAF: major allele frequency; Na: number of alleles; He: expected heterozygosity; PIC: Polymorphic information content; GD:Gene diversity 72 University of Ghana http://ugspace.ug.edu.gh 4.4.2.4. Clustering analysis The dendrogram generated is presented in Figure 4.6. The result revealed three main clusters with six distinct sub-clusters and mean similarity of 0.4. Sub-clusters A and B were found in cluster I, while sub-clusters C and D were found in cluster II. Sub- clusters E and F were found in cluster III. Sub-cluster size varied from 2 to 38 accessions. Sub-clusters, B contained 38 accessions, sub-cluster C contained 20 accessions, sub cluster E contained 18 accessions, sub-clusters A contained 9 accessions and sub- cluster D had 8 of the accessions. Sub-cluster F had the lowest 2 (TR0971 and TR0912) of the genotypes (Table 4.11). 73 University of Ghana http://ugspace.ug.edu.gh A B D Fig. 4. 6: Dendrogram of 96 cassava accessions based on SNP markers 74 University of Ghana http://ugspace.ug.edu.gh Table 4.11: Groupings of the cassava accessions based on SNP markers Cluster I Cluster II Cluster III Sub -cluster A Sub- cluster B Sub-cluster B Sub- cluster C Sub- cluster D Sub-cluster E TR0746 TR0310 TR1791 TR0120 TR1518 TR1696 TR0302 TR0164 TR0575 TR1097 TR0613 TR1036 TR1167 TR1436 TR0142 TR0435 TR0188 TR0075 TR0455 TR0949 TR0779 TR0523 TR0127 TR0329 TR0275 TR0092 TR1175 TR0382 TR0064 TR1162 TR0038 TR1005 TR0024 TR0345 TR0297 TR1035 TR0657 TR0006 TR0432 TR0667 TR1288 TR0519 TR0256 TR0218 TR0159 TR1769 TR0821 TR1041 TR1154 SLICASS4 TR1069 TR0545 TR0864 TR0768 TR0105 TR0300 TR0488 TR0263 TR1326 TR0593 TR1198 TR1827 TR0255 TR0812 TR0358 TR0915 TR0724 TR0590 TR1782 TR0417 TR1523 TR0454 TR0288 TR0135 TR0820 TR0453 TR0356 TR0428 TR0813 TR1716 TR0043 TR1618 TR0694 TR0740 SLICASS6 TR1796 TR0204 TR0589 TR1028 TR1736 Sub- clusterF TR1824 TR0171 TR0971 TR1013 TR0912 75 University of Ghana http://ugspace.ug.edu.gh 4.4.3. Determination of cassava starch quality 4.4.3.1 Chemical properties of cassava starch for 96 genotypes Table 4.12 shows summary statistics for starch quality traits. Average sugar content was 2.18%, with a ranged from 1.19% to 3.92%. There was an asymmetrical frequency distribution, particularly in the case of sugar content (skewness 0.49). Average starch content was 77.68%, with a skewness value of -0.18, the average for water absorption capacity was 101.31 mpas, with a range of variation from 68.31 to 127.82 mpas, with a tendency of values to concentrate towards the higher values (skewness -0.98); the average for amylose content was 24.53% with skewness of -1.59. The average amylopectin value were 75.47 showed skewness value of 1.59, with a range of 72.44 to 82.60 respectively.There were significant differences among the 96 genotypes for all the traits at (P< 0.001). Table 4.12: Summary statistics of starch quality traits from 96 cassava genotypes Parameters SU% SCT% WBC AMY% AYLP Minimum 1.19 58.31 68.31 17.40 72.44 Maximum 3.92 89.89 127.82 27.56 82.60 Average 2.28 77.68 101.31 24.53 75.47 Std. Dev 0.47 7.69 10.42 1.73 1.73 skewness 0.49 -0.18 -0.98 -1.59 1.59 count 96 96 96 96 96 P< value <.001 <.001 <.001 <.001 <.001 SU= sugar %, SCT= starch content %, WBC= water absorption capacity, AMY= amylose content% and AMYP= amylopectin 76 University of Ghana http://ugspace.ug.edu.gh 4.4.3.2.Pasting properties of cassava starch from 96 genotypes Result for pasting properties of cassava are shown in Table 4.13. The average peak viscosity was 343.4 with range of 200.7 to 537.1% with skewness at 0.37. Breakdown viscosity value ranged from 98.8 to 284.7 mPa s with an average around 188.7 mPa s with skewness at -0.01. Average setback viscosity was 66.1 mPa s with a minimum of 25.1 and a maximum value of 120.2 mPa s with a skewness value of 0.37. Final viscosity averaged at 220.8.mPa s, with a range of 142.4 to 368.2 mPa s with skewness value at 1.10. Peak time ranged from 3.5 to 5.3 min with an average of 4.2 min with skewness at 1.5. Average pasting temperature was 74.1oC and ranged from 70.3 to 78.5 with skewness at 0.15.There were significant differences among the cassava genotypes for all the traits at P< 0.001. Table 4.13: Summary statistics for pasting properties from starches of 96 cassava genotypes Parameters PV BD FV ST PT PST Minimum 200.7 98.8 142.4 25.1 3.9 70.3 Maximum 537.1 284.7 368.2 120.2 5.3 78.5 Average 343.4 188.7 220.8 66.1 4.2 74.1 Std Dev 60.3 38.9 40.9 15.5 0.16 1.2 skewness 0.37 -0.01 1.10 0.37 1.5 0.15 count 102 102 102 102 102 102 P>>END cultivate Cassava C2 Do you or any of your 1=yes HH member 2=no cultivate […] cassava within the last 5 years? /___/ /___/ C3 Name the two (2) local and Improve varieties of cassava Variety Names mostly grown in this HH if any? C4 How long have you been growing the main varietiy C5 Iscassava the main crop 1=yes of cultivation for 2=no this HH? C6 If no, what is the main crop C7 Average yearly farm 1=< 1 acre; 2=1-2 acres; 3=2-3 size for […] acres; cassava farm for 4=3-4acres; 5=4-5 acres; 6=>5 this HH? acres /___/ /___/ C8 Main purpose for 1=HH consumption; 2= selling; growing [….] 3=processing to other cassava varieties in this products 4=Consumption & HH? Selling /___/ /___/ C9 Main source of planting 1=Personal farms; 2=Other materials for your farmers;3= Research station; [….] varieties for 4=Commercial farms; this HH? 5=MAFFS; 6=NGOs /___/ /___/ 167 University of Ghana http://ugspace.ug.edu.gh C10 Means of acquiring 1=own materials 2=Purchasing planting materials 3=Gift in this HH for your 3=Exchanging [….] cassava varieties? /___/ /___/ C11 What type of cropping 1= mono-cropping 2=mixed system do you cropping 3= Intercropping mostly usefor cultivating your [….] cassava variety? C12 If mix cropping what is 1=Rice 2= Sorghum 3=Maize the main 4=Groundnut cropincluded for 5=Potatoes 6=Vegetables 7=Pigeon [….]? peas 8=others /___/ /___/ C13 What is the average Yield (kg/acre) yield per year for your most favorite [….]cassava variety grown? C14 Dry mattercontent of 1= High, 2= Moderate, 3= low 4= [….] variety don’t know /___/ /___/ C14 Starch yield of the [….] 1= High, 2= Moderate, 3= low 4= variety don’t know /___/ /___/ SECTION D Household Perception for a New Cassava Varieties D1 Have you or any HH 1=yes member ever 2=no cultivatedhighyielding,s tarch and dry matter cassava? /___/ D2 If yes, What is the name of that variety? Variety Name____________________________ D3 Cassava variety type? 1= Local 2= Improve /___/ D4 What is the main source of 1=Personal farms; 2=Other farmers; 3= the planting material for Research station; 4=Commercial this HH? farms; 5=MAFFS; 6=NGOs /___/ D5 Means of acquiring planting 1=own materials 2=Purchasing 3=Gift materials in this HH? 3=Exchanging /___/ D6 Main purpose for 1=HH consumption; 2= selling; growingthis cassava 3=processing to other cassava variety? products 4=Consumption & Selling /___/ D7 How long have you been growing the main variety 168 University of Ghana http://ugspace.ug.edu.gh D8 If no, would you like to 1=yes grow such a variety in 2=no this HH? /___/ D9 If no, why? E1 FARMERS’ PARTICIPATORY SELECTION OF CASSAVA GENOTYPES High Branching Root size Starch Dry matter Root yield patter conte content/ taste(s n nt mealy weet E1 FARMERS’ PARTICIPATORY SELECTION OF CASSAVA GENOTYPES Pests Diseases Pound- Maturity ability SWOT Analysis No. SWOT Options Tick Rank What are your F1 STRENGHT What are your F2 WEAKNESS What are your F3 OPPORTUNI TIES 169 University of Ghana http://ugspace.ug.edu.gh What are your F4 THREATS Household cassava processors SECTION A Location Information No. NAME 0F LOCATION CODES A1 Questionnaire Identification /___/___/___/___/___/___/___/_ __/___/ A2 Region /___/ Name__________________________________ ______ A3 District /___/___/ Name__________________________________ ______ A4 Chiefdom /___/___/___/___/ Name__________________________________ ____ A5 Village/town Name: ____________________________________ A6 Name of /___/___/ Enumerator_____________________________ __ A7 Date: /___/___/ /___/___/ /___/___/___/___/ A8 Start time: /___/___/ /___/___/ End time: /___/___/ /___/___/ Altitude/hei A9 GPS coordinates of N(S) E(W) ght Industry/Processing unit /___/___/./___/ /___/___/./___/__ /___/___/__ ___/ _/ _/ SECTION B Respondent Information B1.1 Name of respondent B1.2 Age Years /___/___/ 0= female B1.3 Gender 1=male /___/ 170 University of Ghana http://ugspace.ug.edu.gh 0=none ; 1=literate/Koranic ; 2=primary, 3=junior high school, 4=senior high school, B1.4 Level of Education 5=tertiary, 6= other (specify) /___/ 1=married, 2=single, 3=widow/widower, 4=divorced ; 5=Minor (not in B1.5 Marital status age) ; 6=other /___/ Are you the head of the processing 1= Yes, B1.6 centre? 2= No /___/ If no, what isthename and the contact B 1.7 of head? 1= Gari 5=HQCF 2 = starch 6 = Tho B 1.8 What cassava product do you process? 4= fufu 1= Gari 5=HQCF How long have you be processing 2 = starch 6 = Tho B 1.9 cassava product 4= fufu 1 = Daily, 2= Weekly, 3= How often do you process your Monthly 4=quarterly B 1.0 product 5=seasonal How many workers do you have in B1.11 this unit Total number of staff/labour /___/___/ B 1.2 How many Males B1.13 How manyfemale /___/___/ B1.14 Youths/children’s People with age 8 and below /___/___/ 1= Credit facility2=Market B1.15 What is the major benefit derived facility 3=Good processing from the unit facility SECTION C PROCESSING AND PROCESSING CONDITIONS What cassava based product do you 1=starch ,2=Gari ,3=fufu ,4 C1 produce HQCF 5= Others /___/___/ How many tons/kg(….) of the products C2 do you process To make 100kg of the name product, from your experience, what will it cost C3 /___/ 1=starch ,2=Gari ,3=fufu ,4 HQCF To process50kg {...}what quantity of 5=Tho cassava do you use? Do you have varietal preference for 1= Yes, C4 each of the product you process? 2= No /___/ If yes,what is the name of the most suitable cassava variety for C5 processing Name of the variety /___/ 171 University of Ghana http://ugspace.ug.edu.gh 1=local 2=improved C6 Is the variety local or improved /___/ 1= home consumption,2= sale, 3= both Why do you prefer this variety for C7 processing? /___/ 1=Traditional (specify) What processing facilities method do 2=Mechanical (specify) you use? 3=Both C8 /___/ What is the main source of cassava for processing? 1= own production2= from other farmers 3= from market C9 /___/ How do you transport tubers to your 1=Hiring of vehicle 2=on foot 3= processing centre? own vehicle 4= Bikes C10 /___/ SECTION D Preferred cassava and/or cassava based on product characteristics Choose among the5 following characteristics for the type that you prefer most (the least preferred should be at the bottom of the table). Characteristic of (code) Code for preference Dry starch Tubers Gari Cassava flour fufu matte r 1st most preferred 2nd most preferred 3rd most preferred 4th most preferred 5th most preferred Characteristic code: 1= tuber2= dry matter,3= good for making Gari, 4= fufu 5= flour, 6= starch SWOT Analysis No. SWOT Options Tick What are your F1 STRENGHT 172 University of Ghana http://ugspace.ug.edu.gh What are your F2 WEAKNESS What are your F3 OPPORTUNI TIES What are your F4 THREATS Cassava Traders SECTION A Location Information No. NAME 0F LOCATION CODES A1 Questionnaire Identification /___/___/___/___/___/___/___/_ __/___/ A2 Region /___/ Name__________________________________ ______ A3 District /___/___/ Name__________________________________ ______ A4 Chiefdom /___/___/___/___/ Name__________________________________ ____ A5 Village/town Name: ____________________________________ A6 Name of /___/___/ Enumerator_____________________________ __ A7 Date: /___/___/ /___/___/ /___/___/___/___/ A8 Start time: /___/___/ /___/___/ End time: /___/___/ /___/___/ 173 University of Ghana http://ugspace.ug.edu.gh No. NAME 0F LOCATION CODES A1 Questionnaire Identification /___/___/___/___/___/___/___/_ __/___/ Altitude/hei A9 GPS coordinates of N(S) E(W) ght Industry/Processing unit /___/___/./___/ /___/___/./___/__ /___/___/__ ___/ _/ _/ Codes Region District Chiefdom SECTION B Respondent Information B1.1 Name of respondent B1.2 Age Years /___/___/ 0= female B1.3 Gender 1=male /___/ 0=none ; 1=literate/Koranic ; 2=primary, 3=junior high school, 4=senior high school, B1.4 Level of Education 5=tertiary, 6= other (specify) /___/ 1=married, 2=single, 3=widow/widower, 4=divorced ; 5=Minor (not B1.5 Marital status in age) ; 6=other /___/ 1= Yes, Are you the owner of the cassava B1.5 business 2= No /___/ Cassava Product Trading B2 Do you sell any cassava products 1=yes 2=no /___/ 1=cassava root 2 = starch 3 Gari 4=fufu 5=Tho 6 = flour 7= B2.1 If yes, which of the product you sell Leaves /___/ 1= for income,2= home B2.2 If yes why do you sell? consumption 3= both /___/ 174 University of Ghana http://ugspace.ug.edu.gh Which percentage of the cassava products selected above do you B2.3 sell in a growing season /___/ 1=Company 2 =consumers 3=NGO 4=Government B 2.4 Who buys your cassava products agency 5= Other /___/ 1= Yes, Is it difficult to sell your cassava B 2.5 produce? 2= No /___/ 1=Bad road and high cost of transportation If Yes, what is the major reason for the 2= Glut(excess) 3= poor storage lack of market for the cassava you 5= activities of middlemen produce? 6= other B2.6 /___/ Do you make profits from(Starch,Gari,Flour,Fufu,Tube r and Tho)selling the cassava products? B2.7 1= Yes,2= No /___/ 1= starch 2=Gari 3=cassava root 4= others B2.8 Do you sell any of the following /___/ cassava products How often do you market your produce? ( Starch,Gari,Flour,Fufu,Tuber and Tho)) B2.9 /___/ 1= Daily 2= Weekly 3=Monthly 1=periodic markets 2=Daily village/town markets B2.10 3 = Processing centre 4 = street /___/ Where is your main selling location 5 = own product 6 = other Starch trading B3.1 Have you ever bought/sold cassava /___/ starch 1= Yes,2= No 1=periodic markets 2=Daily village/town markets 3 = Processing centre 4 = street /___/ If yes, main buying Location 5 = own product 6 = other B3.2 1= Processor 2= wholesaler 3= /___/ Main suppliers for the products trader 4 =own product 175 University of Ghana http://ugspace.ug.edu.gh B3.3 1=consumer 2=wholesaler /___/ Main buyers type 3=other traders 5 =others 1=periodic markets 2=Daily village/town markets B3.4 3 = Processing centre 4 = street /___/ Main selling location 5 = own product 6 = other Who fixed the prices of starch you sold? B3.5 1= N/A 2= Yourself 3= the buyer 4= Government 1= N/A 2= used prices in neighbouring markets; 3= I used published prices in the newspapers 4= used prices announced on the If prices were fixed by you, say how media5= used cost of you determined them production 6=others B3.6 1= On-farm 2=. At local market ,3= Transported to other parts of Country4= How do you market the cassava Exported to other countries starch produced? 5= Consumed B3.6 Gari Trading B4.1 Have you ever bought/sold cassava /___/ Gari 1= Yes,2= No 1=periodic markets 2=Daily village/town markets 3 = Processing centre 4 = street /___/ If yes, main buying Location 5 = own product 6 = other 1= Processor 2= wholesaler 3= /___/ Main suppliers for the products trader 4 =own product 1=consumer 2=wholesaler /___/ Main buyers type 3=other traders 5 =others 1=periodic markets 2=Daily village/town markets 3 = Processing centre 4 = street /___/ Main selling location 5 = own product 6 = other 176 University of Ghana http://ugspace.ug.edu.gh Consumer SECTION A Location Information No. NAME 0F LOCATION CODES A1 Questionnaire Identification /___/___/___/___/___/___/___/_ __/___/ A2 Region /___/ Name__________________________________ ______ A3 District /___/___/ Name__________________________________ ______ A4 Chiefdom /___/___/___/___/ Name__________________________________ ____ A5 Village/town Name: ____________________________________ A6 Name of /___/___/ Enumerator_____________________________ __ A7 Date: /___/___/ /___/___/ /___/___/___/___/ A8 Start time: /___/___/ /___/___/ End time: /___/___/ /___/___/ Altitude/hei A9 GPS coordinates of N(S) E(W) ght Industry/Processing unit /___/___/./___/ /___/___/./___/__ /___/___/__ ___/ _/ _/ SECTION B Respondent Information B1.1 Name of respondent B1.2 Age Years /___/___/ 0= female B1.3 Gender 1=male /___/ 0=none ; 1=literate/Koranic ; 2=primary, 3=junior high school, 4=senior high school, B1.4 Level of Education 5=tertiary, 6= other (specify) /___/ 177 University of Ghana http://ugspace.ug.edu.gh 1=married, 2=single, 3=widow/widower, 4=divorced ; 5=Minor (not in B1.5 Marital status age) ; 6=other /___/ 1= Yes, B1.6 Are you the head of the household? 2= No /___/ B1.7 Household size Total number of house hold size /___/___/ Total number of males including B1.8 How many Males youths in HH /___/___/ B1.9 Youths/children’s People with age 8 and below /___/___/ SECTION C Consumption Characteristics C1. Do you buy the following cassava starch -products from market Tubers, Gari ,starch ,Fufu? (1 =yes 2= no) C2. Do you recognize when buying, if the cassava are loTubers, Gari ,starch , Fufucal or imported? 1 = Yes, 2 = No C4.If yes, list three major criteria that allow you to recognize the cassava or cassava products if they are local or improved. (Code: 1=price, 2=color, 3=good flavour, 5= good taste7= good for making gari, 8= good for making flour, 9= good for making starch) Tubers /___//___//___/ Gari /___/ /___//___/ Starch /___//___//___/ Fufu /___//___//___/ C5.Are these cassava products always available in the market where you usually go to buy food? (If no, precise the period of scarcity) Cassava products Availability Period of scarcity 1=yes 2=no 1 Cassava starch 2 Raw cassava roots 3 Gari 178 University of Ghana http://ugspace.ug.edu.gh C6. Are your needs in cassava starch products met all through the year? 1 = Yes ___________ 2 = No ____________ C6.1 If no, reasons? C6.2 If yes, why? ECTION DSOURCE OF CASSAVA (MARKETS) D1. Do you produce some of your cassava products yourself? 1= Yes ________ 2= No ________ D2. If yes, how do you use it? 1= Sell ______ 2=consumption ______ 3= both ________ D3. Whatkind of market do you usually go to buy various cassava products for your family? 1= Periodic village/town markets2= Daily village/town markets 3= Road side market 4= Street markets5=Supermarkets THANK YOU!!! 179 University of Ghana http://ugspace.ug.edu.gh APPENDIX 2: Mean performances of 102 cassava accessions for yield related traits Accessions DM NSR HI %ST AYPP COCO 36.33 8.33 0.28 29.23 0.27 SLICASS4 30.33 70.67 0.46 23.27 2.14 SLICASS6 34.33 47.67 0.43 27.40 1.39 TR0006 29.33 59.33 0.41 22.23 1.16 TR0017 28.67 65.33 0.40 21.57 0.89 TR0024 30.33 69.33 0.78 23.30 3.45 TR0038 33.00 73.33 0.53 25.97 2.06 TR0043 31.67 72.00 0.50 24.73 1.73 TR0064 33.00 75.33 0.52 25.90 1.70 TR0075 28.67 44.00 0.47 21.53 1.93 TR0092 30.00 89.33 0.39 22.87 1.85 TR0105 36.67 69.00 0.46 29.70 1.89 TR0120 35.67 63.00 0.26 28.70 0.83 TR0127 34.00 41.33 0.36 26.90 1.09 TR0135 33.33 43.33 0.56 26.37 1.27 TR0142 25.33 44.67 0.25 18.40 1.27 TR0159 35.33 61.67 0.40 28.20 1.24 TR0164 28.33 63.33 0.49 21.30 1.52 TR0171 35.67 47.33 0.53 28.60 1.56 TR0188 32.33 52.67 0.31 25.37 1.35 TR0204 26.00 81.67 0.45 18.87 2.17 TR0218 31.33 52.00 0.47 24.37 1.68 TR0255 35.33 52.33 0.38 28.30 1.13 TR0256 30.67 63.33 0.55 23.70 1.41 TR0263 25.67 54.67 0.31 18.63 0.92 TR0275 28.33 24.00 0.22 21.27 0.27 TR0288 32.33 77.00 0.53 25.37 1.73 TR0297 33.33 60.33 0.24 26.30 0.93 TR0300 23.33 70.33 0.33 16.33 1.54 TR0302 30.00 21.33 0.25 22.83 0.42 TR0310 32.33 39.33 0.33 25.37 1.51 TR0329 32.33 52.33 0.33 25.23 0.91 TR0345 28.33 66.00 0.44 21.27 1.40 TR0356 23.33 85.00 0.45 20.87 1.69 TR0358 30.00 66.33 0.40 23.10 1.47 TR0382 27.67 42.33 0.42 20.67 0.83 TR0417 32.00 69.33 0.38 25.00 0.82 TR0428 30.67 47.67 0.47 23.63 2.08 TR0432 29.00 59.00 0.48 21.97 1.69 TR0435 26.00 42.00 0.43 18.97 1.62 TR0453 30.67 69.67 0.64 23.63 2.40 TR0454 35.67 73.33 0.50 28.63 1.51 TR0455 37.00 62.67 0.51 30.03 1.64 TR0488 34.00 67.00 0.48 27.00 2.23 TR0519 26.33 47.67 0.52 19.17 1.53 TR0523 33.00 84.00 0.64 25.87 2.32 TR0545 24.00 102.00 0.47 16.97 1.88 TR0575 28.00 37.33 0.48 20.83 0.88 TR0589 27.33 60.67 0.47 20.20 1.57 TR0590 29.67 94.00 0.63 22.73 3.14 TR0591 31.67 65.33 0.39 24.67 1.45 TR0593 33.67 42.67 0.49 26.57 1.06 TR0613 34.00 52.67 0.47 26.87 1.94 TR0657 30.67 65.67 0.33 23.57 1.30 180 University of Ghana http://ugspace.ug.edu.gh Cont’d mean performance of yield and it related traits Accessions DM NSR HI %ST AYPP TR0667 35.00 76.33 0.62 28.00 2.74 TR0694 23.33 20.33 0.33 16.40 1.00 TR0724 33.33 53.67 0.45 26.27 1.31 TR0740 27.00 38.00 0.43 19.97 0.77 TR0745 31.00 69.67 0.49 23.97 1.76 TR0746 33.00 101.67 0.47 25.87 2.33 TR0768 28.00 73.33 0.53 21.00 2.26 TR0779 29.67 41.33 0.47 22.67 0.98 TR0812 35.00 57.00 0.48 27.87 1.17 TR0813 31.00 53.33 0.46 23.93 1.43 TR0820 32.33 69.67 0.48 25.20 1.61 TR0821 29.00 55.67 0.48 22.07 1.78 TR0864 31.00 70.67 0.61 23.87 2.79 TR0912 29.33 67.67 0.40 22.37 1.37 TR0915 34.67 78.00 0.48 27.57 1.89 TR0949 36.00 50.00 0.54 29.07 1.53 TR0971 32.67 43.67 0.53 25.53 1.46 TR0991 31.67 74.33 0.38 24.60 1.08 TR1005 32.33 38.00 0.39 25.23 0.75 TR1013 31.67 42.33 0.20 24.67 0.52 TR1028 26.33 42.00 0.20 19.27 0.60 TR1035 32.00 97.33 0.49 24.90 1.46 TR1036 26.00 63.00 0.44 18.90 1.58 TR1041 34.33 39.33 0.38 27.27 0.97 TR1069 27.33 32.00 0.73 20.23 1.57 TR1097 28.67 52.67 0.46 21.60 1.39 TR1154 32.33 75.00 0.54 25.17 2.37 TR1159 31.33 58.33 0.49 24.30 1.83 TR1162 32.33 83.33 0.35 25.17 1.72 TR1167 30.33 85.67 0.73 23.27 2.71 TR1175 34.33 60.33 0.58 27.33 1.99 TR1198 28.00 57.00 0.58 20.90 1.69 TR1288 32.33 47.67 0.50 25.33 1.82 TR1326 30.33 49.67 0.43 23.20 1.45 TR1436 31.00 71.67 0.31 23.93 1.56 TR1518 30.67 61.00 0.42 23.60 1.81 TR1523 31.33 96.33 0.50 24.27 2.38 TR1618 31.67 52.00 0.70 24.67 2.70 TR1696 32.67 34.33 0.44 25.60 1.37 TR1716 32.33 91.33 0.45 25.27 1.29 TR1736 31.33 44.00 0.52 24.23 2.11 TR1769 31.67 65.67 0.55 24.63 2.43 TR1776 28.67 83.67 0.53 21.70 2.49 TR1782 23.00 47.00 0.46 16.13 2.00 TR1791 30.33 52.00 0.37 23.23 0.94 TR1796 30.67 41.67 0.40 23.60 0.96 TR1824 31.33 66.00 0.37 24.33 1.35 TR1827 35.00 59.67 0.28 27.90 0.95 MEAN 30.89 59.77 0.45 23.89 1.56 CV% 12.8 59.8 0.5 16.8 32.7 SE 5.3 21.2 0.1 5.2 0.7 P-value <.0001 <.0001 <.0001 <.0001 <.0001 181 University of Ghana http://ugspace.ug.edu.gh APPENDIX 3 :Mean performance of leaf characterise traits Accessions LL PL WLB PH NLL COCO 11.67 12.57 2.97 145.00 7.00 SLICASS 4 10.13 19.93 2.87 158.33 6.33 SLICASS 6 9.97 12.03 2.97 163.67 4.33 TR 0006 10.10 11.17 3.73 139.33 6.00 TR 0017 12.87 14.40 2.23 147.33 6.33 TR 0024 11.10 14.67 2.87 148.67 6.33 TR 0038 9.47 11.90 2.80 165.00 5.00 TR 0043 10.63 10.97 5.87 169.67 5.67 TR 0064 10.47 18.73 6.53 140.33 7.67 TR 0075 10.03 14.87 2.70 158.33 6.67 TR 0092 11.13 13.50 7.90 162.00 7.00 TR 0105 13.17 18.50 3.20 155.00 5.67 TR 0120 8.10 10.60 2.30 149.67 5.67 TR 0127 13.70 16.77 3.80 160.33 6.67 TR 0135 10.20 10.83 2.97 140.33 6.33 TR 0142 10.50 9.10 3.23 173.00 5.00 TR 0159 10.03 12.53 7.23 136.33 4.33 TR 0164 12.07 14.93 7.93 157.33 5.67 TR 0171 8.17 10.13 5.03 133.00 5.67 TR 0188 10.37 11.67 7.43 141.00 6.33 TR 0204 8.65 12.55 5.40 155.00 7.00 TR 0218 10.57 7.60 11.53 162.67 6.33 TR 0255 10.27 17.47 6.70 147.33 5.00 TR 0256 8.63 13.67 5.00 125.00 4.33 TR 0263 13.73 12.33 3.50 128.33 7.00 TR 0275 10.53 8.93 3.27 170.67 4.33 TR 0288 12.07 16.03 10.07 158.33 6.33 TR 0297 10.03 13.40 7.87 159.33 6.33 TR 0300 8.47 11.33 2.57 147.00 5.67 TR 0302 9.43 12.03 2.57 145.00 5.67 TR 0310 10.67 10.17 8.40 147.00 6.33 TR 0329 12.07 15.60 9.00 159.67 7.00 TR 0345 11.23 12.87 4.60 155.00 5.67 TR 0356 9.73 11.80 5.13 134.00 5.00 TR 0358 12.37 14.93 7.87 163.33 7.33 TR 0382 8.87 14.50 2.23 154.67 6.33 TR 0417 9.80 7.47 2.83 127.00 5.67 TR 0428 11.57 15.80 3.5 156.67 6.33 TR 0432 9.97 13.60 2.37 142.33 5.67 TR 0435 12.77 17.87 3.63 158.00 6.33 TR 0453 9.10 10.40 2.10 159.67 5.33 TR 0454 11.83 14.23 2.73 144.00 6.33 TR 0455 11.33 11.57 5.700 154.67 5.00 TR 0488 11.23 14.23 3.83 147.33 5.67 TR 0519 9.77 9.97 2.57 150.33 5.00 TR 0523 11.60 15.17 3.00 150.33 5.67 TR 0545 12.53 17.53 3.33 150.33 7.00 TR 0575 9.77 17.33 7.73 136.00 4.67 TR 0589 10.50 9.54 4.90 119.33 5.00 TR 0590 11.17 15.13 7.2 130.33 6.33 TR 0591 11.23 13.97 7.03 168.67 7.00 TR 0593 12.70 13.9 9.2 154.67 5.00 TR 0613 11.53 18.53 7.63 135.67 5.67 182 University of Ghana http://ugspace.ug.edu.gh Cont’d mean performance of leaf traits Accessions LL PL WLB PH NLL TR 0657 9.37 12.20 2.37 154.67 5.67 TR 0667 10.53 20.17 2.60 125.67 7.00 TR 0694 10.97 7.43 7.90 164.67 6.33 TR 0724 11.67 14.37 2.77 153.67 5.67 TR 0740 8.10 16.05 9.00 98.00 5.67 TR 0745 9.93 18.90 2.83 124.67 5.00 TR 0746 10.03 14.83 2.57 164.00 7.00 TR 0768 9.53 13.03 4.97 145.33 4.33 TR 0779 8.20 12.50 2.40 172.67 4.67 TR 0812 9.67 15.03 2.27 138.67 4.33 TR 0813 13.30 15.70 5.60 142.00 5.67 TR 0820 9.67 14.37 2.63 152.67 5.00 TR 0821 11.37 16.27 3.27 148.67 5.67 TR 0864 15.53 14.87 5.43 142.33 6.33 TR 0912 11.83 10.67 4.63 158.33 6.33 TR 0915 8.93 13.60 3.17 144.00 4.33 TR 0949 8.83 9.47 3.80 148.67 4.33 TR 0971 11.40 10.30 2.97 159.00 5.67 TR 0991 9.97 14.37 2.50 129.00 3.00 TR 1005 11.60 14.53 1.70 153.67 7.00 TR 1013 9.27 12.50 6.13 155.00 5.00 TR 1028 11.57 9.40 3.60 155.67 5.67 TR 1035 8.87 8.20 1.87 139.00 6.33 TR 1036 10.30 14.07 3.60 161.67 5.67 TR 1041 9.70 13.97 3.77 148.67 5.00 TR 1069 9.70 13.77 2.70 121.00 5.00 TR 1097 10.93 11.67 2.37 165.67 7.67 TR 1154 10.30 13.63 3.80 180.00 6.33 TR 1159 10.33 13.13 2.67 152.33 6.33 TR 1162 9.30 13.70 2.73 160.00 6.67 TR 1167 10.70 10.07 6.03 162.67 5.00 TR 1175 12.33 18.77 3.03 132.00 6.33 TR 1198 7.53 9.67 2.53 152.67 5.33 TR 1288 10.53 17.37 6.27 145.67 6.33 TR 1326 9.90 14.33 2.23 139.00 5.00 TR 1436 12.13 9.40 6.93 163.00 5.67 TR 1518 11.13 14.53 2.60 134.00 6.33 TR 1523 12.67 19.37 7.13 158.33 6.33 TR 1618 11.47 11.63 4.10 149.67 5.67 TR 1696 10.73 12.53 2.47 138.33 6.00 TR 1716 9.97 12.63 2.30 139.00 5.67 TR 1736 10.37 11.40 3.43 163.33 5.00 TR 1769 8.97 15.43 2.73 158.33 6.33 TR 1776 12.07 13.17 6.43 162.67 6.00 TR 1782 10.90 11.20 3.33 128.67 7.00 TR 1791 10.47 18.23 2.53 136.33 5.33 TR 1796 9.70 16.60 2.53 154.00 7.00 TR 1824 13.10 15.67 5.97 166.67 6.33 TR 1827 9.47 11.30 5.93 130.33 6.00 TR0204 12.30 11.40 2.80 156.00 7.00 Mean 10.64 13.5 4.39 149.17 5.83 CV% 11.4 20.9 100.4 17.55 55.1 SE 2.4 5.2 4.2 25.7 1.5 P-value <0.0001 <0.0001 <0.9707 <0.3193 <0.0098 183 University of Ghana http://ugspace.ug.edu.gh Proximate analysis for starch chemical properties among 96 cassava starch samples Gen SU SC WAC AMY AMP COCO 2.63±0.02a 67.25±0.21e 102.67±2.65cd 24.02±0.39a 75.98±0.39d SLICASS4 2.48±0.05a 88.93±0.00c 103.19±0.04cd 24.88±0.39a 75.12±0.39d SLICASS6 2.42±0.04a 77.42±0.39c 126.22±1.09a 24.96±0.00a 75.04±0.00d TR0006 2.58±0.02a 88.28±0.10c 85.35±3.96abc 26.10±0.12a 73.89±0.12d TR0017 2.31±0.05a 85.16±0.10c 98.87±0.06cd 24.02±0.39a 75.98±0.39d TR0024 2.17±0.00a 82.08±0.10c 105.44±0.43cd 24.53±0.04a 75.47±0.04d TR0038 2.96±0.95a 85.74±0.10c 99.65±1.34cd 25.24±0.04a 74.76±0.04d TR0043 2.06±0.05a 89.84±0.01c 97.96±7.92cd 25.04±0.08a 74.96±0.09d TR0064 2.06±0.03a 83.08±0.10c 104.24±1.77cd 25.59±0.39a 74.41±0.39d TR0075 1.43±0.01a 74.84±0.16c 97.74±1.19cd 26.02±0.04a 73.98±0.04d TR0092 2.27±0.01a 70.72±0.03c 106.52±4.43cd 24.84±0.93a 75.16±0.93d TR0105 3.06±0.17a 70.99±0.38c 75.87±1.52cd 25.59±0.39a 74.41±0.39d TR0120 2.73±0.05a 70.01±0.21c 107.57±4.01cd 22.83±0.78a 77.17±0.79d TR0127 1.67±0.05a 86.39±0.31c 96.08±3.18cd 23.27±0.35a 76.73±0.35d TR0135 1.98±0.04a 76.31±0.39c 104.92±2.51cd 23.15±0.24a 76.85±0.24d TR0142 2.17±0.02a 73.59±0.29c 104.99±3.60cd 23.23±0.39a 76.77±0.39d TR0157 2.20±0.03a 80.66±0.49c 82.21±1.26cd 23.23±0.39a 76.77±0.39d TR0188 2.57±0.01a 86.00±0.32c 103.39±3.29cd 24.84±0.01a 75.16±0.01d TR0204 2.93±0.18a 86.49±0.42c 105.21±3.19cd 24.02±0.47a 75.98±0.47d TR0218 3.15±0.02de 79.11±0.20c 107.62±0.19cd 24.29±0.04a 75.70±0.04d TR0255 1.70±0.01a 58.37±0.06bc 101.78±0.41cd 17.68±0.28d 82.32±0.28a TR0263 2.55±0.01a 76.94±0.39c 103.59±0.71cd 24.09±0.09a 75.90±0.08d TR0265 2.04±0.04a 76.61±0.38c 105.89±0.21cd 26.18±0.19a 73.82±0.19d TR0275 2.53±0.06a 85.49±0.21c 108.86±0.50cd 23.62±0.57a 76.38±0.57d TR0288 3.10±0.05a 81.65±0.30c 105.59±2.06cd 24.41±0.08a 75.59±0.08d TR0297 2.44±0.01a 71.75±0.38c 109.32±1.07cd 24.41±0.79a 75.59±0.79d TR0300 2.34±0.01a 67.12±0.86a 104.11±4.41cd 25.71±0.12a 74.29±0.12d TR0302 2.05±0.05a 72.63±0.21c 75.29±0.98cd 25.19±0.08a 74.80±0.08d TR0310 2.18±0.03a 88.48±0.02c 102.37±0.03cd 24.02±0.18a 75.98±0.18d TR0345 2.07±0.05a 78.17±0.00c 87.99±1.09cd 23.54±0.87a 76.46±0.87d TR0356 2.01±0.05a 77.83±0.10c 111.57±4.74cd 25.19±0.00a 74.80±0.00d TR0358 2.54±0.07a 80.79±0.10c 99.92±7.89cd 24.84±0.43a 75.16±0.43d TR0382 2.54±0.05a 74.51±0.11c 101.13±8.70cd 23.98±0.43a 76.02±0.43d 184 University of Ghana http://ugspace.ug.edu.gh Cont’d proximate analysis for starch chemical properties among 96 cassava starch samples Gen SU SC WAC AMY AMP TR0428 2.84±0.05a 71.68±0.21c 106.04±0.55cd 25.71±0.04a 74.29±0.04d TR0432 3.32±0.02a 74.51±0.28c 103.59±0.82cd 25.39±0.04a 74.61±0.04d TR0453 2.18±0.00a 86.01±0.74c 96.145±0.66cd 25.98±0.00a 74.02±0.00d TR0454 2.51±0.05a 72.46±0.03c 103.093±1.72cd 25.59±0.39a 74.41±0.39d TR0455 2.18±0.03a 69.98±0.37c 104.014±1.19cd 25.87±0.04a 74.13±0.04d TR0488 2.45±0.02a 89.01±0.43c 96.06±0.88cd 23.23±0.39a 76.77±0.39d TR0519 2.23±0.00a 69.39±0.21c 101.57±0.83cd 24.61±0.04a 75.39±0.04d TR0523 2.30±0.03a 87.65±0.32c 106.69±1.053cd 26.81±0.04a 73.19±0.04d TR0545 2.45±0.03a 85.89±0.42c 100.71±0.792cd 26.81±0.04a 73.19±0.04d TR0575 2.07±0.04a 70.25±0.03c 103.28±3.24cd 22.40±0.51a 77.59±0.51d TR0589 1.83±0.06a 68.19±0.04c 115.98±3.87bc 24.80±0.39a 75.19±0.39d TR0590 2.53±0.04a 89.46±0.43c 72.28±3.97cde 25.79±0.51a 74.21±0.51d TR0591 2.18±0.01a 73.14±0.15c 107.39±2.54cd 23.62±0.00a 76.38±0.00d TR0593 2.30±0.04a 76.25±0.02c 80.39±1.39cd 22.44±0.18a 77.56±1.18d TR0631 2.74±0.04a 72.87±0.39c 106.51±0.04cd 24.88±0.25a 75.12±1.26d TR0657 1.72±0.04a 86.43±0.41c 79.92±3.70cd 24.92±0.43a 75.08±0.43d TR0667 2.37±0.02a 73.59±0.38c 106.22±2.09cd 25.59±0.08a 74.41±0.08d TR0694 1.88±0.23a 74.22±2.27c 107.01±2.04cd 22.40±0.04a 77.59±0.04d TR0724 1.73±0.04a 69.68±0.13abc 103.07±3.17cd 22.16±0.04a 77.83±0.04d TR0740 1.97±0.03a 83.53±4.21c 100.86±0.89cd 26.10±0.04a 73.89±0.04d TR0745 2.39±0.02a 67.68±0.10c 102.99±1.26cd 23.23±0.39a 76.77±0.34d TR0746 2.96±0.47a 89.19±0.43c 96.05±2.00cd 23.62±1.57a 76.38±0.57d TR0768 2.35±0.05a 71.82±0.07c 102.69±2.25cd 25.12±0.63a 74.88±0.63d TR0778 3.27±0.09ab 76.74±0.25c 107.02±2.39cd 23.62±0.00a 76.38±0.00d TR0812 2.02±0.06a 80.66±0.49c 79.44±5.27cd 26.10±0.04a 73.89±0.04d TR0813 3.09±0.04a 72.28±0.38c 79.18±4.46cd 23.70±0.08a 76.29±0.08d TR0820 2.33±0.02a 76.39±9.06c 103.95±5.02cd 26.77±0.00a 73.23±0.00d TR0821 2.04±0.04a 72.92±0.38c 107.90±0.36cd 23.89±0.12a 76.10±0.12d TR0864 1.28±0.03a 59.04±0.45cd 74.22±5.42cd 24.92±0.43a 75.08±0.43d TR0912 2.02±0.07a 65.32±0.61de 104.41±1.55cd 21.18±0.16cd 78.82±0.16bc TR0915 1.77±0.02a 73.77±0.21c 109.16±7.61cd 24.80±0.26a 75.19±0.26d TR0949 1.95±0.16a 83.40±0.42c 103.91±2.69cd 23.74±0.04a 76.26±0.04d 185 University of Ghana http://ugspace.ug.edu.gh Cont’d proximate analysis for starch chemical properties among 96 cassava starch samples Gen SU SC WAC AMY AMP TR0971 2.28±0.06a 80.59±0.79c 102.28±2.14cd 25.63±0.35a 74.37±0.35d TR0991 2.049±0.01a 86.16±0.51c 109.86±0.58cd 25.67±0.39a 74.33±0.39d TR1002 2.02±0.06a 85.56±0.11c 103.90±3.45cd 24.02±0.39a 75.98±0.39d TR1013 2.53±0.02a 87.03±0.11c 109.54±3.04cd 26.46±0.00a 73.54±0.00d TR1028 2.19±0.05a 87.75±0.01ab 103.04±1.36cd 24.69±0.28a 75.31±0.28d TR1035 2.75±0.02a 68.49±0.10c 107.95±1.25cd 24.41±0.00a 75.59±0.00d TR1036 2.21±0.02a 66.04±0.20cde 106.62±1.06cd 24.02±0.39a 75.98±0.39d TR1041 2.06±0.04a 77.24±4.72c 109.99±0.36cd 24.02±0.39a 75.98±0.39d TR1097 1.49±0.05a 87.03±0.11c 107.59±7.68cd 24.80±0.39a 75.19±0.39d TR1154 1.67±0.01a 85.04±0.61c 96.90±5.35cd 23.66±0.83a 76.34±0.83d TR1159 2.02±0.02a 64.90±0.16c 76.16±1.33cd 23.56±0.02a 76.44±0.02d TR1162 1.87±0.01a 75.92±0.01c 97.55±4.20cd 23.62±3.15a 76.38±0.15d TR1167 3.12±0.02a 89.65±0.24c 124.66±1.42ab 23.98±0.43a 76.02±0.43d TR1175 2.02±0.05a 72.69±0.10c 103.34±2.32cd 25.04±0.08a 74.96±0.08d TR1198 1.92±0.01a 85.25±0.00c 110.82±1.74cd 24.80±0.39a 75.19±0.39d TR1288 1.46±0.05a 78.70±0.83c 95.21±4.16cd 26.02±0.04a 73.98±0.04d TR1326 1.84±0.05a 74.44±4.73c 102.10±2.27cd 26.18±0.38a 73.82±0.38d TR1436 2.39±0.06a 76.07±0.17c 104.20±1.90cd 25.24±0.04a 74.76±0.04d TR1518 3.22±0.11abc 86.36±0.51c 107.04±1.01cd 23.62±0.00a 76.38±0.00d TR1523 2.62±0.02a 74.81±0.08c 99.79±0.74cd 24.96±0.39a 75.04±0.39d TR1618 3.18±0.15abcd 78.60±0.01c 100.06±0.02cd 24.41±0.79a 75.59±0.79d TR1697 1.23±0.05a 79.71±0.10c 113.65±6.42cd 25.98±0.00a 74.02±0.00d TR1716 2.37±0.05a 80.48±0.10c 90.44±1.02cd 26.89±0.04a 73.11±0.04d TR1736 2.31±0.00a 71.41±0.10c 106.61±1.01cd 24.80±0.08a 75.19±0.08d TR1769 2.14±0.06a 77.54±0.48c 98.75±1.31cd 24.80±1.18a 75.19±0.18d TR1776 2.27±0.04a 82.79±0.41c 113.39±3.99cd 23.50±0.04a 76.49±0.04d TR1782 2.79±0.01a 74.44±0.16c 104.29±0.07cd 26.38±0.39a 73.62±0.39d TR1791 1.88±0.22a 85.39±.44c 106.33±2.25cd 26.10±0.04a 73.89±0.04d TR1796 2.15±0.05a 69.40±0.20c 103.25±0.74cd 27.17±0.39a 72.83±0.39d TR1824 1.76±0.04a 75.11±0.47c 96.85±1.23cd 24.01±0.39a 75.98±0.39d TR1827 3.10±0.02a 82.06±0.51c 104.31±0.89cd 19.33±0.04cd 80.67±0.04ab CV% 7.63 3.46 4.29 4.47 1.55 Mean 2.28 77.68 101.31 24.52 74.15 LSD (5%) 0.33 4.27 8.57 0.18 0.18 Values were calculated from two replicates; ± standard deviation. Results followed by the same superscript in a column were not significantly different at P<0.05.Gen=genotypes; SU = Sugar content; SC=Starch content: AMY = amylose content; AMP = amylopectin; 186 University of Ghana http://ugspace.ug.edu.gh APPENDIX 4: ANOVA result for GXE analysis for different location Combined ANOVA for number of storage root Source DF Type III SS Mean Square F Value Pr > F REP(LOCATION) 6 11926.5897 1987.7650 5.56 <.0001 LOCATION 2 116426.9060 58213.4530 162.84 <.0001 GEN 25 33810.2607 1352.4104 3.78 <.0001 LOCATION*GEN 50 91178.6496 1823.5730 5.10 <.0001 Combined ANOVA for harvest index Source DF Type III SS Mean Square F Value Pr > F REP(LOCATION) 6 0.01192689 0.00198782 0.09 0.9969 LOCATION 2 3.38904956 1.69452478 79.70 <.0001 GEN 25 0.71864682 0.02874587 1.35 0.1375 LOCATION*GEN 50 1.15324314 0.02306486 1.08 0.3474 Combined ANOVA for fresh storage root yield Source DF SS M S F V Pr > F REP(LOCATION) 6 327.561882 54.593647 0.75 0.6102 LOCATION 2 6588.809572 3294.404786 45.27 <.0001 GEN 25 2752.039959 110.081598 1.51 0.0683 LOCATION*GEN 50 7302.337406 146.046748 2.01 0.0007 187 University of Ghana http://ugspace.ug.edu.gh Combined ANOVA for starch yield Source DF SS M S F V Pr > F REP(LOCATION) 6 156.473590 26.078932 1.27 0.2728 LOCATION 2 2677.519060 1338.759530 65.38 <.0001 GEN 25 903.862607 36.154504 1.77 0.0201 LOCATION*GEN 50 903.963162 18.079263 0.88 0.6892 Combined ANOVA for dry matter content Source DF SS M S F V Pr > F REP(LOCATION) 6 327.561882 54.593647 0.75 0.6102 LOCATION 2 6588.809572 3294.404786 45.27 <.0001 GEN 25 2752.039959 110.081598 1.51 0.0683 LOCATION*GEN 50 7302.337406 146.046748 2.01 0.0007 Combined ANOVA cassava mosaic disease Source DF SS M S F V Pr > F REP(LOCATION) 6 1.58974359 0.26495726 0.31 0.9329 LOCATION 2 81.87179487 40.93589744 47.33 <.0001 GEN 25 87.67948718 3.50717949 4.05 <.0001 LOCATION*GEN 50 99.46153846 1.98923077 2.30 <.0001 188