Agamah et al. Malar J (2021) 20:421 https://doi.org/10.1186/s12936-021-03955-0 Malaria Journal RESEARCH Open Access Network-driven analysis of human– Plasmodium falciparum interactome: processes for malaria drug discovery and extracting in silico targets Francis E. Agamah1,2, Delesa Damena1, Michelle Skelton2, Anita Ghansah3, Gaston K. Mazandu1,2,4* and Emile R. Chimusa1* Abstract Background: The emergence and spread of malaria drug resistance have resulted in the need to understand disease mechanisms and importantly identify essential targets and potential drug candidates. Malaria infection involves the complex interaction between the host and pathogen, thus, functional interactions between human and Plasmodium falciparum is essential to obtain a holistic view of the genetic architecture of malaria. Several functional interaction studies have extended the understanding of malaria disease and integrating such datasets would provide further insights towards understanding drug resistance and/or genetic resistance/susceptibility, disease pathogenesis, and drug discovery. Methods: This study curated and analysed data including pathogen and host selective genes, host and pathogen protein sequence data, protein–protein interaction datasets, and drug data from literature and databases to perform human-host and P. falciparum network-based analysis. An integrative computational framework is presented that was developed and found to be reasonably accurate based on various evaluations, applications, and experimental evidence of outputs produced, from data-driven analysis. Results: This approach revealed 8 hub protein targets essential for parasite and human host-directed malaria drug therapy. In a semantic similarity approach, 26 potential repurposable drugs involved in regulating host immune response to inflammatory-driven disorders and/or inhibiting residual malaria infection that can be appropriated for malaria treatment. Further analysis of host–pathogen network shortest paths enabled the prediction of immune- related biological processes and pathways subverted by P. falciparum to increase its within-host survival. Conclusions: Host–pathogen network analysis reveals potential drug targets and biological processes and pathways subverted by P. falciparum to enhance its within malaria host survival. The results presented have implications for drug discovery and will inform experimental studies. Keywords: Malaria, Drug resistance, Genomics, Multi-omics, Gene ontology, Protein–protein interaction Background Plasmodium falciparum malaria is a common infec- *Correspondence: gmazandu@gmail.com; emile.chimusa@uct.ac.za 1 Division of Human Genetics, Department of Pathology, Institute tious disease in Africa, and arguably the most important of Infectious Disease and Molecular Medicine, Faculty of Health Sciences, parasitic disease in the world, posing a significant pub- University of Cape Town, Cape Town, South Africa lic health burden as compared to other World Health Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/l icen ses/ by/4. 0/. The Creative Commons Public Domain Dedication waiver (http:// creat iveco mmons. org/ publi cdoma in/z ero/1. 0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Agamah et al. Malar J (2021) 20:421 Page 2 of 20 Organization (WHO) disease-endemic regions. For polymorphisms of known drug-resistance genes, such as instance, Africa contributed to about 93% (213 million of pfcrt, pfmdr1, pfk13, pfmrp1, pfdhfr, and pfdhps generally 228 million) and 94% (380,000 of 405,000) of global cases express effects that counteract drugs controlling the dis- and deaths, respectively in 2018 [1]. ease [7, 10–12]. Compared to the clinical phenotype of The use of anti-malarial drugs has been the opti- resistance to quinolones and SP which usually takes the mal avenue for controlling the disease. Currently, arte- form of reduced accumulation of drugs within the para- misinin-based combination therapy (ACT) is used as site, particularly targets, artemisinin resistance, manifests the first-line option for malaria treatment globally [2]. as slow parasite clearance in patients and is characterized ACT was adopted in Africa after the decline in efficacy by the parasite’s ability to alter intraerythrocytic cell cycle of previous widely used anti-malarial drugs, including with an increased ring stage and a shortened trophozoite chloroquine and sulfadoxine-pyrimethamine (SP) [2]. stage [8, 13]. This was to ensure that, each component of the combi- Falciparum malaria is a multifactorial disease that natorial drug acts through different mechanisms within involves the complex interplay between the host, vector, the parasite, aiming to significantly reduce the likeli- and the pathogen [14, 15]. The host–pathogen interac- hood of the emergence of multi-drug resistant parasites. tions have been a driving selective force influencing the Unfortunately, the parasite has shown tremendous ability genetic architecture of both species, particularly, on how to develop resistance and tolerance to these artemisinin their genes are involved in drug and/or genetic resist- derivatives and the long half-life partner drugs in some ance, disease susceptibility, and the infection processes countries of the Greater Mekong Sub-region [2–4]. With [14, 16, 17]. several reports supporting parasite recrudescence and a Understanding these interactions requires an in-depth significant decrease in their sensitivity to ACT, there has analysis of the organism’s proteome which is regarded been continuous surveillance to monitor the emergence to execute the genetic programme. Proteins execute and spread of artemisinin-resistant parasite strains in functions mostly through extended networks with each Africa and elucidate whether it will follow a similar pat- other thereby forming a framework of the sensitive and tern observed for chloroquine and SP resistance where complex regulatory system underlying a wide degree of resistant strains originated from Southeast Asia [2, 4–7]. post-translational modifications and processes [18]. The Interestingly, a study conducted by Uwimana et  al. [7] complex physicochemical dynamic connections formed has demonstrated the independent emergence and local within the system facilitate the structural and functional spread of artemisinin partial resistance in Rwanda driven organization of the organism. These connections make by R561H mutation in kelch gene. Another study con- up the protein–protein interaction network (PPIN). ducted in Northern Uganda has also reported independ- Recent advances in host and parasite genomics in ent emergence and local spread of artemisinin-resistant terms of high-throughput proteomics studies, host and parasite driven by mutations in the A675V or C469Y parasite genome sequencing have led to a corresponding allele in the kelch13 gene [8]. These pieces of evidence increase in biological datasets that describe the transi- suggest that artemisinin resistance has emerged indepen- tion of species over time, particularly, the metabolic and dently in Eastern Africa. developmental stages of pathogens. As such, the applica- Researchers have proposed that the emergence of arte- tion of computational approaches to efficiently mine the misinin parasite-resistant strains in Africa would result inter and intra-species functional interactions to address in about 78 million additional cases [4] and over 100,000 the challenges presented by the disease is critical [19]. A deaths annually [9]. Evidence abounds to the fact that a systematic and comprehensive study of these complex major challenge to controlling, eliminating, and eradicat- interactions is essential in elucidating relevant pathways, ing malaria is drug resistance. It is the principal reason signalling, drug resistance patterns, genes-gene products for the expansion of this life-threatening disease. inter-relationships, and drug targets as well as developing The architectural framework of the parasite’s genome novel hypotheses and models to predict disease causality constitutes a major framework influencing variations [20]. in the levels of the drug susceptibility, particularly hav- In this study, a network-based integrative compu- ing elucidated that P. falciparum anti-malarial drug tational framework was leveraged to predict protein resistance involves a single major gene effect. Sponta- targets that may be used to guide the rational design neous alterations in the form of single nucleotide vari- of pathogen- and host-directed therapies for malaria ation and multiple mutations in different genes within treatment. Following the target prediction, a seman- the parasite genome capacitate the pathogen’s ability to tic similarity approach was implemented to prioritize develop tolerance mechanisms or resist the drug action informed potentially repurposable drugs that can be over time thus, yielding the unexpected result. Genetic engineered for malaria treatment. Further analysis of A gamah et al. Malar J (2021) 20:421 Page 3 of 20 host–pathogen network shortest paths enabled the pre- 3D7 strain), respectively. Genes with no correspond- diction of immune-related biological processes and path- ing UniProt protein ID as at the time of this study were ways potentially subverted by P. falciparum to increase discarded. its within-host survival. Human malaria susceptibility-associated single nucleo- tide polymorphisms (SNPs) were retrieved from GWAS Methods summary statistics datasets obtained from Malaria- Study design and procedures GEN [21]. The summary statistics dataset comprised Various open access heterogeneous genomic and func- of 20,273,529 spanning across chromosome one (1) to tional datasets retrieved from databases and literature twenty-two (22). In this study, approximately 690,000 using text mining techniques were used as inputs for significant SNPs (p-value < 0.05) were filtered for further analysis. The approach for this study (Fig. 1) consisted of analysis. These SNPs were then mapped onto 44 genes five main steps: (1) data curation and pre-processing, (2) (herein referred to as host candidate genes, Additional scoring and integrating functional datasets; (3) biologi- file 5: Table S2) using the dbSNP annotated data [22, 23]. cal network assembling and structural analysis; (4) gene mapping and enrichment analysis (5) implicit semantic Scoring and integrating functional datasets similarity approaches to predict malaria-similar diseases The study performed pathogen-pathogen, pathogen- and repurposable drugs. Briefly, the framework uses inte- host, and host-host protein sequence BLAST using their grative, scoring, and clustering algorithms coupled with respective protein sequences retrieved from the UniProt statistical methods and biological knowledge to analyse database [24]. This was followed by implementing an and validate results. information-theoretic based functional scoring scheme outlined by Mazandu and Mulder [25] and summarized Data pre‑processing in the Additional file  10: (Eqs.  1–8) to score the func- The various datasets utilized for this study are described tional associations obtained from sequence BLAST and in Additional file  4: Table  S1. To achieve uniform iden- the conserved domains interaction datasets from the tifiers (IDs) and convenient data manipulation, all genes InterPro database [26]. and protein IDs were mapped to only reviewed proteins from Swiss-Prot under the non-redundant UniProt iden- Scoring high‑throughput experimental datasets tifier system for harmonization. Human and P. falcipa- and interologs rum genes were mapped to UniProt proteins with taxon To incorporate curated functional interaction datasets identifier 9609 and 36,329 (Plasmodium falciparum in the analysis, the following criteria were defined to Fig. 1 An overview of the approach implemented in this study Agamah et al. Malar J (2021) 20:421 Page 4 of 20 prioritize and score pair-wise interactions from experi- interaction reliability or confidence score cut-off are used mental and interolog datasets retrieved from Reactome to reduce the impact of these biases, leading to a PPI net- [27], IntAct [28], MINT [29], BIOGRID [30], and lit- work of high confidence interactions with an increased erature [31–36]. The criteria for scoring were based on; coverage  [37]. Further analyses only used medium and (1) the number of experimental methods that have con- high confidence interactions or interactions predicted by firmed such pair-wise functional interaction, (2) the two different sources. To evaluate the structural features number of databases that have reported such pair-wise of nodes (proteins) and edges (interactions), network functional interaction, and (3) the number of times the centrality metrics including node degree, betweenness, pair-wise functional interaction has been reported in and closeness (Additional file 10: Eqs. 9–11) were com- the literature. For every pair-wise functional interaction puted. High degree nodes with low betweenness describe supported by one evidence, a reliability score of 0.3 was degree-based or ‘local’ subnetwork interconnectiv- assigned, else, a reliability score of 0.7 if it is supported by ity mostly between functionally related proteins. High two or more pieces of evidence. degree nodes with high betweenness contribute to struc- tural-based or ‘global’ subnetwork interconnectivity and Biological network assembling and structural analysis signal transmission thus, promoting system-level func- Table 1 describes the number of proteins retrieved from tional integration. Node closeness describes the average each dataset, the number of reviewed proteins/genes shortest length between neighbouring nodes determin- considered from each input dataset and the pair-wise ing the proximity to information sharing and biological functional interaction implemented for further down- process execution between functionally related nodes stream analysis. From the pre-processed scored datasets, [38]. the functional interactions obtained were categorized, as low (scores less than 0.3), medium (scores ranging Community structure and hub classification between 0.3 and 0.7), and high confidence levels (scores The study aimed to identify hub genes/proteins that greater than 0.7). Biases may exist in the PPI network establish links with multiple functional clusters (com- generated due to relatively high noise related to high- munities), thus, characterized by both ‘local’ and ‘global’ throughput data or experiments from which interactions network interconnectivity, structural, and functional are derived. In the absence of gold standard PPIs, inte- features. To predict the hubs, clustering analysis was grating data from different sources and applying strict performed to identify network communities of densely Table 1 Extracted functional interactions between manually annotated proteins Interaction Number of Number of Number of Number of pair‑wise References source Interactions proteins reviewed interactions between reviewed proteins proteins A. Functional interactions between annotated Plasmodium falciparum proteins LaCount P. falciparum PPI 2864 1308 62 17 [35] Wuchty et al. in silico Plasmodium PPI (1) 19,979 2321 85 74 [31] Wuchty et al. experimental PPI (2) 1428 361 12 11 [32] Wuchty experimental P. falciparum PPI (3) 5458 1986 91 32 [33] Wuchty et al. experimental P. falciparum PPI (4) 4918 1872 81 15 [34] IntAct 2916 1343 67 26 [28] InterPro 1013 256 98 241 [26] Scored BLAST sequence similarity 1090 (BLAST) 163 130 231 [85] STRING 617 163 114 386 [86] B. Functional interactions between annotated human proteins Reactome 79,619 8059 5029 19,736 [27] Score BLAST sequence similarity 3,807,888 20,395 9611 143,533 [85] (BLAST) InterPro 2,646,550 35,928 17,797 231,799 [26] Bossi and Lerner 80,922 10,229 8416 54,238 [36] STRING 11,759,454 19,354 18,836 5,244,655 [86] IntAct 456,263 35,770 16,061 169,627 [28] A gamah et al. Malar J (2021) 20:421 Page 5 of 20 connected nodes using a variant of an integrative com- the Kappa statistic, Jaccard, and the Best Match Average putational algorithm that implements the Blondel et  al. (BMA) measures (Additional file 10). The score is a quan- [39] heuristic method based on modularity optimization. titative measure of the underlying shared biological pro- This clustering model is a scalable hierarchical agglom- cesses among the disease targets. A higher score between erative method based on modularity optimization and disease enriched processes suggests that the disease-pair has been shown to outperform all other known commu- and their associated candidate proteins are functionally nity detection methods [40], including Smart Local Mov- similar thus, the likelihood for similar treatment options. ing [41], Infomap [42], and Label Propagation [43], in A similarity score threshold was defined based on the terms of computation time or complexity and the quality upper quartile and interquartile range of the distribu- of the communities detected (modularity). The parasite tion given by tr = Q3+ ε ∗ IQR , where ε , tr,Q3 and IQR candidate genes (herein referring to known antimalarial represent the tuning parameter (0 ≤ ε ≤ 1.5) threshold, resistant genes and reported genes expressing signature upper quartile, and interquartile range, respectively. of selection towards drug resistance) retrieved from lit- erature [2, 6, 10] and host candidate gene-encoded pro- Results teins (Additional file 5: Table S2) were mapped onto the Network clustering and functional annotation analysis assembled parasite and host networks to cluster the net- The generated parasite network consists of 662 unique works. The subnetworks were explored to identify global interactions among 140 characterized proteins (Fig. 2A). hubs, herein defined as candidate gene/proteins charac- The unified host network assembled comprised of terized by a high degree and high betweenness score. 4,133,136 unique functional interactions between 20,329 nodes. The host-parasite network consisted of 31,512 Functional annotation analysis unique functional interactions between 8023 proteins. Gene annotation and enrichment analysis were per- The topology properties of the generated networks were formed to elucidate statistically significant biological pro- explored to investigate the relationships between the cesses and pathways to which the hub genes are involved. degree, betweenness, and closeness centrality measures. Biological processes were inferred from the gene ontol- As shown in Additional file 1: Fig. S1, subnetworks were ogy database [44], whereas pathway information was classified as either degree-based (subnetworks formed obtained from PlasmoDB v46 [45] and the KEGG data- from nodes with a high degree but low betweenness) or base [46]. By applying the hypergeometric test [47], structural-based (subnetworks formed from nodes with p-values of processes and pathways were estimated, lev- high degree, high betweenness, and high closeness). The eraging on their frequency of occurrence. The Bonferroni nodes forming the degree-based and structural-based multiple correction test [47] was then implemented to subnetworks are herein referred to as key proteins. estimate the adjusted p-values. Network clustering analysis reveals disease candidate key Semantic similarity proteins/genes as hubs The development of human disease ontology terms [48] The purpose of clustering is to partition the complex net- has provided an enriched platform of human disease work into subnetworks and identify essential communi- data to evaluate similarities between various diseases of ties and critical functional nodes. It is a way of grouping different disorder classes based on gene-related molecu- nodes in the network into modules sharing functional lar functions. The analysis is based on the hypothesis connectivity. The parasite network (Fig.  2A) consists of that varying combinations of disease-associated genes 8 clusters of which 5 contained key proteins whereas the can influence the pathogenicity of similar diseases [49]. dense human network consisted of 32 clusters of which To predict repurposable drugs for malaria treatment, 7 contained key proteins. From the network clustering an in-house python-based semantic model was imple- (Additional file  2: Fig. S2A, Additional file  3: Fig. S2B), mented for disease and drug similarity. The model uses two parasite candidate key proteins were identified as host candidate key proteins, disease-target datasets, and hubs, C6KTD2 (SET1) and C6KTB7 (PFF1365c) both gene ontology datasets as input data to make predictions on chromosome 6. These parasite candidate key pro- based on functional similarities inferred from associated teins are involved in the merozoite developmental stage gene ontology terms. The semantic similarity approach where they invade red blood cells (RBCs), cause disease was further implemented to identify diseases that are severity, and contribute to the exponential growth of biologically similar to malaria. In the analysis, the seman- the parasite population [50]. Analysis of the host net- tic similarity score between the pair of diseases was work revealed 6 candidate key proteins as hubs; P22301 leveraged to identify and prioritize diseases similar to (IL10 [MIM: 124092]), P05362 (ICAM1 [MIM: 147840]), malaria. The similarity score was estimated by computing P01375 (TNF [MIM: 191160]), P30480 (HLA-B [MIM: Agamah et al. Malar J (2021) 20:421 Page 6 of 20 Fig. 2 A Assembled parasite network and B Functional interactions between C6KTD2 and C6KTB7 subnetwork within the parasite network. The nodes common to the subnetworks are coloured in yellow 142830]), P16284 (PECAM1 [MIM: 173445]) and O00206 parasitized red blood cells (RBCs), parasite sequestra- (TLR4 [MIM: 603030]). These proteins are cognate tion in organs rupture, and removal of infected RBCs host receptors that respond to inflammation by releas- [50, 51]. Most importantly, the identified host candidate ing pro-inflammatory cytokines, enhancing adhesion of key proteins are targets for drugs in DrugBank [52] and A gamah et al. Malar J (2021) 20:421 Page 7 of 20 have been reported to offer higher opportunities for receptors sense infection. To contribute to this effort, the drug repurposing, although a smaller proportion of the shortest paths between the parasite and host hub pro- human genome is druggable [53–55]. Additional file  6: teins within the host-parasite network were explored to Table S3 and Additional file 7: Table S4 describe the iden- gain insight into the most likely routes for innate immune tified candidate key proteins prioritized by the degree, response interference by the parasite. betweenness, and closeness scores. Studies have shown that the shortest path analysis of a functional network yields high coverage compared to Biological processes and pathway enrichment of hub direct neighbours within the network [56]. The shortest genes path between host–pathogen disease-associated can- The identified hub genes within the subnetworks were didate key genes herein refer to the minimum number used for the functional annotation process. The results of edges required to connect these genes. Longer paths revealed 4 statistically significant essential processes and consist of more nodes (proteins) involved in a cascade of an enriched pathway (Table 2) specific to the parasite key signalling processes to trigger innate immune responses hub genes. A total of 23 significant biological processes by inducing the production of chemokines and cytokines and 21 enriched pathways (Table  3) were identified to upon parasite infection. It is, therefore, a measure of underly host hub gene’s contribution towards malaria information relay between the hub genes thus, the shorter infection. From the host perspective, the hub genes are the path, the quicker the transmission and the relevance mainly involved in immune regulatory biological pro- of the interaction in investigating immune adaptiveness cesses within immune-related pathways (47.6%), parasitic and parasite pathogenesis [56]. It is noteworthy that, disease-related pathways (23.8%), bacteria disease-related shortest path lengths between the pathogen disease-asso- pathways (14.2%), endocrine and metabolic disease- ciated genes and human disease-associated genes con- related pathways (4.7%), viral disease-related pathway ferring immunity in the functional network are the most (4.7%) and transport and catabolism related pathway feasible routes of parasite invasion of host immunity and (4.7%)[44, 46]. Most importantly, the malaria pathway escaping the contribution of host genetics towards drug ranked the most significant pathway with both p-value action [56, 57]. Most importantly, shortest paths would and adjusted p-value of 0. This supports the association trigger excessive activation which may be deleterious as of these hub genes to malaria. The enriched pathways it can cause systemic inflammation and disease [50]. This, presented the likelihood of similarity between malaria therefore, suggests that developing immune-modulatory and other diseases. drugs that target the host targets can induce an immune response to avoid the state of been overwhelmed by the Shortest path analysis between hub genes reveals parasite. functional insights towards disease progression The results showed that the shortest path between The study investigated functional interactions between parasite hub proteins and any of the host hub pro- the host and pathogen targets in the context of parasite teins were between O00206—C6KTB7, and O00206- survival, host immune tolerance, and how it can inform C6KTD2 as shown in Table  4. Such paths were drug discovery research. The immune tolerance machin- characterized by mediators. These mediators are mostly ery remains to be the natural driving force influencing signal receptors involved in cell regulatory activi- the parasite’s survival when host–pathogen recognition ties, production of cytokines, transcription processes, Table 2 Statistically significant biological processes and pathways of key P. falciparum malaria-associated genes inferred from PlasmoDB v46 and gene ontology database Enriched biological process Gene Ontology (GO)‑ID Process Gene ontology term name P‑value Adjusted p‑value GO:0019904 Protein domain specific binding 1.77e−3 7.08e-3 GO:0004842 Ubiquitin-protein transferase activity 5.31e−3 4.25e-2 GO:0019787 Ubiquitin-like protein transferase activity 5.75e−3 4.60e-2 GO:0051568 Histone h3-k4 methylation 0.0103149 0.030945 A. Enriched pathway Pathway ID Pathway-name P-value Adjusted p-value ec00310 Lysine degradation 3.87e-2 3.87e−2 Agamah et al. Malar J (2021) 20:421 Page 8 of 20 Table 3 Statistically significant biological processes and enriched pathways of key human malaria-associated genes inferred from gene ontology and KEGG database Gene ontology Gene ontology term name P‑value Adjusted (GO)‑id process P‑value Enriched biological process Go:0042346 Positive regulation of nfkappab import into nucleus 2.00161e−05 0.00432 Go:0045348 Positive regulation of mhc 2.40637e−06 0.00052 Class ii biosynthetic process Go:0032689 Negative regulation of 3.10034e−05 0.00670 Interferon-gamma production Go:0007157 Heterophilic cell–cell adhesion 0.00012 0.02714 Via plasma membrane cell adhesion molecules Go:2000352 Negative regulation of endothelial 2.70760e−05 0.00585 Cell apoptotic process Go:0032715 Negative regulation of interleukin-6 production 8.99764e−08 1.9434e-05 Go:2000343 Positive regulation of chemokine (c-x-c motif ) ligand 1.68841e−05 0.00364 Go:0032729 Positive regulation of 0.00012 0.02713 Interferon-gamma production Go:0070374 Positive regulation of erk1 2.8883e−05 0.00623 And erk2 cascade Go:0050830 Defence response to gram-positive bacterium 2.65221e−05 0.00572 Go:0034116 Positive regulation of heterotypic cell–cell adhesion 1.68841e−05 0.00364 Go:0044130 Negative regulation of 6.07930e−06 0.00131 Growth of symbiont in host Go:0030198 Extracellular matrix organization 5.39819e−05 0.01166 Go:0045416 Positive regulation of 2.40637e−06 0.00051 Interleukin-8 biosynthetic process Go:0032755 Positive regulation of 0.00016 0.03562 Interleukin-6 production Go:0002740 Negative regulation of cytokine 1.00303e−06 0.00021 Secretion involved in immune response Go:0045429 Positive regulation of nitric 1.23374e−07 2.665e−05 Oxide biosynthetic process Go:0043032 Positive regulation of 1.02165e−05 0.00220 Macrophage activation Go:1904999 Positive regulation of leukocyte 2.00663e−07 4.3343e−05 Adhesion to arterial endothelial Cell Go:0031663 Lipopolysaccharide mediated 2.42641e−08 5.2410e−06 Signalling pathway Go:1904707 Positive regulation of vascular 0.00016 0.03663 Smooth muscle cell proliferation Go:0032800 Receptor biosynthetic process 6.68766e−07 0.00014 Go:1900227 Positive regulation of nlrp3 inflammasome complex assembly 2.70759e−05 0.00584 A. Enriched pathway Kegg pathway id Kegg-pathway-name P-value Adjusted p-value Hsa05144 Malaria 0.0 0.0 Hsa05310 Asthma 6.52960e−07 7.966e−05 Hsa04145 Phagosome 1.46003e−06 0.00017 Hsa05146 Amoebiasis 0.00014 0.01745 Hsa04640 Hematopoietic cell lineage 1.70661e−06 0.00020 Hsa05330 Allograft rejection 3.03329e−06 0.00037 Hsa05133 Pertussis 8.77858e−06 0.00107 Hsa04940 Type i diabetes mellitus 9.00124e−05 0.01098 Hsa05162 Measles 8.93218e−05 0.0108 Hsa04650 Natural killer cell mediated cytotoxicity 0.00014 0.01724 A gamah et al. Malar J (2021) 20:421 Page 9 of 20 Table 3 (continued) Gene ontology Gene ontology term name P‑value Adjusted (GO)‑id process P‑value Hsa04657 Il—17 signalling pathway 0.00037 0.04624 Hsa05152 Tuberculosis 1.92895e−10 2.35332e−08 Hsa05150 Staphylococcus aureus infection 4.40440e−08 5.37336e−06 Hsa05142 Chagas disease (american trypanosomiasis) 7.77032e−05 0.00947 Hsa05143 African trypanosomiasis 7.32790e−10 8.9400e-08 Hsa05140 Leishmaniasis 1.16192e−11 1.41754e−09 Hsa05321 Inflammatory bowel disease (ibd) 9.68744e−08 1.18186e−05 Hsa05322 Systemic lupus erythematosus 3.51918e−05 0.00429 Hsa05323 Rheumatoid arthritis 0.00027 0.03331 Hsa05320 Autoimmune thyroid disease 9.62632e−06 0.00117 Hsa05332 Graft—versus—host disease 0.00010 0.01229 and regulating cell survival and apoptosis. The short- Nucleoprotein TPR (TPR [MIM: 189940]) and Gigax- est paths identified (Table  4) suggest that inhibition onin (GAN [MIM: 605379]). or alteration to the proper functioning of each path might help the parasite to survive immune responses, Predicting repurposable drugs for malaria treatment based thus, the aggregation of small effects. The development on Implicit Semantic Similarity of adaptive immunity is expected to happen when the After defining a semantic similarity score threshold (as parasite undergoes diversity throughout time such that illustrated in Fig.  3A), 1944 (8.04%) out of 24,166 dis- they evade the host system when they become tolerant eases in the DisGeNet platform version 6 were identified and establish different mechanisms to interfere with the to be semantically like malaria. The disease hits were fil- host’s response [58]. These interferences can also be in tered by maintaining those whose targets are involved in the form of the production of effector mechanisms that the same pathways of host Malaria hub genes. The dis- can down-regulate innate immunity [59]. The results ease hits were further filtered by maintaining diseases have shown that the dynamic patterns to parasite sur- supported by biological evidence from the literature. vival and immune adaptiveness are mediated by other The final filtered disease hits consisted of 113 diseases human-specific genes or proteins conferring immunity. (Additional file 8: Table S5). These identified diseases fall Importantly, pfk13 is known to be associated with in the category of infectious, inflammatory, and genetic artemisinin resistance, but little is known of its interac- neurological diseases which trigger the human immune tion with host genes/proteins and how that influences machinery to overproduce cytokines; confirming the fact drug resistance or parasite survival within the host. that malaria is an inflammatory response-driven disease. Further network analysis was performed to explore Among the top disease hits includes sickle cell anae- interactions between pfk13 and the host candidate key mia [MIM: 603903], liver dysfunction [MIM: 613759], proteins. The results revealed no functional interactions fever ([MIM: 142680], [MIM: 614371]), hepatitis ([MIM: between pfk13 and the host hub genes. However, the 606518], [MIM: 609532]) and respiratory distress syn- analysis showed interactions between pfk13 and highly drome [MIM: 267450]. It is interesting to note that the expressed host kelch-like proteins and regulatory genes disease hits described have been reported to be gov- involved in essential processes such as transcription erned by the same pathologic principles as malaria infec- regulation, cell-surface, cell–cell signalling, and regula- tion [60, 61].Finally, to predict repurposable drugs, 1426 tion of phosphorylation. Among the regulatory genes approved drugs and their corresponding targets were include the transcriptional regulator Kaiso (ZBTB33), retrieved from the DrugBank database. Next, non-human Zinc finger and BTB domain-containing protein 17 drugs were excluded and were remained with 1282 drugs (ZBTB17 [MIM: 604084]), BTB/POZ domain-contain- and their targets for further downstream analysis. The ing protein 10 (KCTD10 [MIM: 613421]), Zinc fin- drugs were further filtered to retain those with target pro- ger and BTB domain-containing protein 10 (ZBTB10 cesses associated with malaria and the predicted malaria [MIM: 618576]), Myoneurin (MYNN [MIM: 606042]), similar diseases. Then after, the semantic approach was implemented to predict putative repurposable drugs. Agamah et al. Malar J (2021) 20:421 Page 10 of 20 Table 4 Shortest paths linking O00206 (TLR4) and parasite hub nodes within the host–pathogen unified functional network Host candidate Mediator Mediator gene name [OMIM ID] Mediator description Parasite Potential parasite adaptive biological process key protein candidate key protein A. Shortest paths linking O00206 and C6KTB7 (SET1) nodes within the host–pathogen unified functional network O00206 Q9BYH8 NFKBIZ [MIM: 608004] NF-kappa-B inhibitor zeta C6KTB7 Inflammatory response; T cell receptor signalling pathway O00206 Q05823 RNASEL [MIM: 180435] 2-5A-dependent ribonuclease C6KTB7 Interferon alpha/beta signalling, positive regulation of transcription by RNA polymerase II O00206 Q5S007 LRRK2 [MIM: 609007] Leucine-rich repeat serine/threonine-protein kinase 2 C6KTB7 Activation of MAPK activity; O00206 Q38SD2 LRRK1 [MIM: 610986] Leucine-rich repeat serine/threonine-protein kinase 1 C6KTB7 Positive regulation of intracellular signal transduction O00206 O75762 TRPA1 [MIM: 604775] Transient receptor potential cation channel subfamily A C6KTB7 Cell surface receptor signalling pathway member 1 O00206 Q96HA7 TONSL [MIM: 604546] Tonsoku-like protein C6KTB7 Cytoplasmic sequestering of transcription factor O00206 Q00653 NFKB2 [MIM: 164012] Nuclear factor NF-kappa-B p100 subunit C6KTB7 Inflammatory response; innate immune response; NIK/ NF-kappaB signalling; negative regulation of transcription by RNA polymerase II O00206 P25963 NFKBIA [MIM: 164008] NF-kappa-B inhibitor alpha C6KTB7 Positive regulation of inflammatory response; apoptotic process; I-kappaB kinase/NF-kappaB signalling O00206 P46531 NOTCH1 [MIM: 190198] Neurogenic locus notch homolog protein 1 C6KTB7 Immune response O00206 P20749 BCL3 [MIM: 109560] B-cell lymphoma 3 protein C6KTB7 Regulation of apoptotic process; regulation of interferon- gamma production; T-helper 1 type immune response; positive regulation of interferon-gamma production O00206 P42771 CDKN2A [MIM: 600160] Cyclin-dependent kinase inhibitor 2A C6KTB7 Apoptotic process; O00206 Q8NI38 NFKBID [MIM: 618887] NF-kappa-B inhibitor delta C6KTB7 Inflammatory response; innate immune response; O00206 P19838 NFKB1 [MIM: 164011] Nuclear factor NF-kappa-B p105 subunit C6KTB7 Apoptotic process; inflammatory response; innate immune response; regulation of transcription by RNA polymerase II O00206 Q8NDB2 BANK1 [MIM: 610292] B-cell scaffold protein with ankyrin repeats C6KTB7 B cell activation; positive regulation of interleukin-6 production O00206 Q15653 NFKBIB [MIM: 604495] NF-kappa-B inhibitor beta C6KTB7 Signal transduction B. Shortest paths linking O00206 and C6KTD2 within the host–pathogen unified functional network O00206 Q13114 TRAF3 [MIM: 601896] TNF receptor-associated factor 3 C6KTD2 Apoptotic process; innate immune response; toll-like receptor signalling pathway; regulation of interferon-beta production; positive regulation of JNK cascade; O00206 Q86WT6 TRIM69 [MIM: 616017] E3 ubiquitin-protein ligase TRIM69 C6KTD2 Apoptotic process O00206 Q12933 TRAF2 [MIM: 601895] TNF receptor-associated factor 2 C6KTD2 Positive regulation of JNK cascade; apoptotic process; O00206 Q9UPN9 TRIM33 [MIM: 605769] E3 ubiquitin-protein ligase TRIM33 C6KTD2 Negative regulation of transcription by RNA polymerase II; O00206 P55895 RAG2 [MIM: 179616] V(D)J recombination-activating protein 2 C6KTD2 B cell homeostatic proliferation; O00206 O75626 PRDM1 [MIM: 603423] PR domain zinc finger protein 1 C6KTD2 Adaptive immune response; innate immune response; negative regulation of transcription by RNA polymerase II; O00206 Q96CA5 BIRC7 [MIM: 605737] Baculoviral IAP repeat-containing protein 7 C6KTD2 Apoptotic process Agamah et al. Malar J (2021) 20:421 Page 11 of 20 Table 4 (continued) Host candidate Mediator Mediator gene name [OMIM ID] Mediator description Parasite Potential parasite adaptive biological process key protein candidate key protein O00206 Q8NHM5 KDM2B [MIM: 609078] Lysine-specific demethylase 2B C6KTD2 Negative regulation of transcription by RNA polymerase II; apoptotic process; O00206 Q9BUZ4 TRAF4 [MIM: 602464] TNF receptor-associated factor 4 C6KTD2 Apoptotic process; regulation of I-kappaB kinase/NF- kappaB signalling; O00206 Q96P53 WDFY2 [MIM: 610418] WD repeat and FYVE domain-containing protein 2 C6KTD2 Positive regulation of protein phosphorylation O00206 P15918 RAG1 [MIM: 179615] V(D)J recombination-activating protein 1 C6KTD2 adaptive immune response; immune response; O00206 P19474 TRIM21 [MIM: 109092] E3 ubiquitin-protein ligase TRIM21 C6KTD2 Innate immune response; response to interferon-gamma O00206 Q9Y4K3 TRAF6 [MIM: 602355] TNF receptor-associated factor 6 C6KTD2 Activation of MAPK activity; positive regulation of JNK cascade; apoptotic process; toll-like receptor signalling pathway; regulation of transcription by RNA polymerase II O00206 Q6PCT2 FBXL19 [MIM: 609085] F-box/LRR-repeat protein 19 C6KTD2 Post-translational protein modification O00206 Q14258 TRIM25 [MIM: 600453] E3 ubiquitin/ISG15 ligase TRIM25 C6KTD2 Innate immune response; O00206 Q9UNE7 STUB1 [MIM: 607207] E3 ubiquitin-protein ligase CHIP C6KTD2 Protein ubiquitination O00206 Q15075 EEA1 [MIM: 605070] Early endosome antigen 1 C6KTD2 Endocytosis O00206 Q8IWB7 WDFY1 [MIM: 618080] WD repeat and FYVE domain-containing protein 1 C6KTD2 Positive regulation of toll-like receptor 3 and 4 signalling pathway O00206 Q09472 EP300 [MIM: 602700] Histone acetyltransferase p300 C6KTD2 Apoptotic process; positive regulation of NIK/NF-kappaB signalling O00206 Q8IYM9 TRIM22 [MIM: 606559] E3 ubiquitin-protein ligase TRIM22 C6KTD2 positive regulation of I-kappaB kinase/NF-kappaB signal- ling O00206 Q9Y2K7 KDM2A [MIM: 605657] Lysine-specific demethylase 2A C6KTD2 Histone H3-K36 demethylation O00206 Q9NQV6 PRDM10 [MIM: 618319] PR domain zinc finger protein 10 C6KTD2 Positive regulation of transcription by RNA polymerase II O00206 Q92793 CREBBP [MIM: 600140] CREB-binding protein C6KTD2 Positive regulation of transcription by RNA polymerase II; apoptotic process O00206 Q6UWE0 LRSAM1 [MIM: 610933] E3 ubiquitin-protein ligase LRSAM1 C6KTD2 Ubiquitin-dependent endocytosis; O00206 Q8WVD3 RNF138 [MIM: 616319] E3 ubiquitin-protein ligase RNF138 C6KTD2 Protein ubiquitination O00206 Q8TCQ1 MARCHF1 [MIM: 613331] E3 ubiquitin-protein ligase MARCHF1 C6KTD2 Immune response O00206 Q6Q0C0 TRAF7 [MIM: 606692] E3 ubiquitin-protein ligase TRAF7 C6KTD2 Apoptotic process; positive regulation of MAPK cascade O00206 O00463 TRAF5 [MIM: 602356] TNF receptor-associated factor 5 C6KTD2 Apoptotic process; positive regulation of I-kappaB kinase/ NF-kappaB signalling; positive regulation of JNK cascade; O00206 Q13233 MAP3K1 [MIM: 600982] Mitogen-activated protein kinase kinase kinase 1 C6KTD2 Activation of protein kinase activity O00206 P98170 XIAP [MIM: 300079] E3 ubiquitin-protein ligase XIAP C6KTD2 Regulation of innate immune response; regulation of inflammatory response; apoptic process O00206 Q6ZMZ0 RNF19B [MIM: 610872] E3 ubiquitin-protein ligase RNF19B C6KTD2 Adaptive immune response O00206 O14964 HGS [MIM: 604375] Hepatocyte growth factor-regulated tyrosine kinase C6KTD2 Regulation of MAP kinase activity; positive regulation of substrate gene expression O00206 Q9NWF9 RNF216 [MIM:] E3 ubiquitin-protein ligase RNF216 C6KTD2 Apoptotic process Agamah et al. Malar J (2021) 20:421 Page 12 of 20 Fig. 3 A Different distributions of disease similarity scores obtained in terms of frequencies (proportions) of disease matches vs similarity scores between disease-associated processes. The bigger rectangular bar indicates the threshold for the similarity between disease pairs of which the enriched similarity score (ESS) were used for further analysis. B Distributions of drug similarity scores obtained in terms of the relative frequency of drug matches against functional similarity scores between candidate gene and drug. The bigger rectangular bar indicates the threshold for the similarity between drug pairs of which the enriched similarity score (ESS) were used for further analysis From the identified drugs sharing some similarities in involved in, the results revealed 26 potential repurpos- terms of processes, those that are over 1.5 of the inter- able drugs (Additional file 9: Table S6).The repurposable quartile range were extracted and ordered. With a drugs categorized as known anti-malarial, monoclonal defined similarity score threshold of 0.31099875 (Fig. 3B) antibodies, immunomodulators, herbs, natural products, based on similarity in terms of processes the drugs are Janus kinase inhibitors, and thrombolytic agents act as A gamah et al. Malar J (2021) 20:421 Page 13 of 20 either antagonist, agonists, inhibitors, or precursors tar- nodes (candidate key proteins) that contribute signifi- geting genes over-represented in immune response and cantly to the stability and integrity of the network. Gene cytokine-mediated signalling processes. Janus kinase annotation and enrichment analysis of the identified inhibitors including ruxolitinib, are known for their abil- hub genes were performed to elucidate underlying sta- ity to effectively inhibit the production of cytokines and tistically significant biological processes and pathways. cause eryptosis contributing to the clearance of eryth- Also, shortest paths analysis was performed to elucidate rocytes infected with malaria, decreased parasitaemia, pathways that could account for parasite adaptiveness and protection against severe malaria [62]. The results to host response and potential drug resistance develop- showed that drugs involved in regulating host immune ment. From the parasite assembled functional network, response to inflammatory-driven disorders target the the analysis performed predicted C6KTD2 (SET1) and Tumour necrosis factor and inhibit its activity to regulate C6KTB7 (PFF1365c) as key targets. These targets are downstream processes such as pro-inflammatory cascade essential at specific developmental stages of the parasite signalling. Several of the potentially repurposable drugs and have been reported as candidates for drug and vac- are used for treating some diseases like malaria including cine development. The results confirm the importance of rheumatoid arthritis, ischemic stroke, psoriatic arthritis, these targets. Also, the analysis (Figs. 2B and 4A) showed and idiopathic arthritis. that these targets could be critical for combinatorial The drug hits include chloroquine, infliximab, hydroxy- drug design. There is an accumulation of evidence that chloroquine, glucosamine, ginseng, minocycline, rux- C6KTB7 is a potential multi-stage target for a malaria olitinib, and natalizumab which can be appropriated for vaccine and drug development [64–68]. C6KTB7 is malaria treatment. These drug hits have been reported to mainly involved in ubiquitin-protein transferase activity control malaria infection by inhibiting residual malaria (GO:0004842, GO:0019787) through the protein ubiqui- infection, knocking parasite gene expression, and acti- tination and modification pathway (UPA00143). Studies vating eryptosis. Furthermore, some of the hits such as have shown that many biological processes and substrates adalimumab, Natalizumab, etanercept, thalidomide, are targeted by the ubiquitin pathway such that instability ustekinumab, and canakinumab are anti-TNF monoclo- or modification in ubiquitination and deubiquitination nal antibodies and anti-inflammatory agents that could reactions influences the pathogenesis of many eukaryotic modulate the immune response to severe and cerebral system-related diseases [65]. For instance, the dysregula- malaria. The analysis also predicted thrombolytic agents tion of ubiquitin ligase is associated with neurodegenera- such as anistreplase, reteplase, alteplase, and tenect- tive disorders, such as Parkinson’s disease and infectious eplase which can play an essential role in the treatment diseases including tuberculosis [66]. This is usually asso- of coagulopathy in malaria, particularly among severe ciated with interference with immune response. C6KTB7 and cerebral malaria infections [63]. Considering malaria significantly influences the parasite’s development and as an inflammatory-response driven disease presenting malaria pathogenesis by regulating various cellular pro- with multiple manifestations, these putative drug hits can cesses and pathways critical for the pathogen’s survival undergo both computational and experimental reposi- in the human host [69]. This phenomenon usually hap- tioning for adjunctive malaria therapy, particularly severe pens as a result of post-translational modifications within and cerebral malaria. the biological system through processes such as tran- scriptional regulation and cell cycle progression [66]. Discussion For example, the protein is responsible for the positive In this study, an integrative network-based framework regulation of DNA-templated transcription and epige- was implemented on the various heterogeneous experi- netic factors such as histone H3-K4 methylation, essen- mental and in silico datasets retrieved from databases tial for transcription regulation [65]. Interestingly, studies and literature to assemble Plasmodium falciparum, have shown that inhibition of the activities of C6KTB7 human, and human-Plasmodium falciparum functional and the ubiquitin–proteasome system has the poten- protein–protein interaction network. Using host-malaria tial for many disease treatments including P. falciparum GWAS summary statistics datasets, host-disease-asso- malaria [65, 68, 69]. Of note, the parasite candidate pro- ciated genes were identified by mapping nominally sig- teins are essential during specific developmental stages. nificant SNPs to their associated genes. The identified For instance, Aminake et al. [68] explored the role of the genes, malaria parasite selective variants, and parasite proteasome of P. falciparum for malaria drug research variants under strong signature of selection were mapped and revealed C6KTB7 as a component of the ubiquitin– onto the host and pathogen functional network respec- proteasome which could serve as a promising multi-stage tively to identify key subnetworks. The subnetworks of (liver, blood, and transmission stages of the pathogen) each assembled network were evaluated to investigate target, thus a supporting results presented by Chung Agamah et al. Malar J (2021) 20:421 Page 14 of 20 et al. [70]. Additionally, Ponts et al. [65] showed that pro- apoptotic processes. Positive regulation of NIK/NF-kap- teins involved in the ubiquitylation pathway including paB signalling (GO:0042346) process responsible for the the ubiquitin ligases (E3) such as C6KTB7 (PFF1365c) regulation of NF-kappaB importation has been studied to influence parasite virulence, thus targeting such a path- be involved in immune and inflammatory responses, par- way may represent new therapeutic targets for apicom- ticularly in eukaryotic cells. Down or negative regulation plexan parasites, such as P. falciparum. This suggests of NF-kappaB has been reported to be associated with that inhibiting parasite adaptation to the ubiquitylation P. falciparum-modulated endothelium transcriptome pathway and the proteins involved (including putative E3 contributing to cerebral malaria [79]. Positive regulation ubiquitin-protein ligase protein PFF1365c (C6KTB7)) is of the MHC class II biosynthetic process (GO:0045348) important for malaria drug research [65, 68]. C6KTD2 process has been shown to regulate immune response to is a possible candidate for effective malaria vaccine malaria [80]. Pre-erythrocytic immunity to malaria (cer- development [67]. The protein plays an essential role in ebral malaria) is linked to MHC antigens such that vari- chromatin structure, protein domain-specific binding. ations in class I and class II in these antigens contribute and gene expression in the parasite [35, 71]. Also, it is significantly to malaria susceptibility thus, reduced, or mainly involved in the histone lysine methylation post- increased host immune response [80]. Also, other pro- translational modification process (GO: 0051568) which cesses such as negative regulation of interferon-gamma usually involves the synergistic effect of histone-lysine production (GO:0032689), negative regulation of inter- methyltransferases and histone lysine demethylases [71, leukin-6 production (GO:0032715), negative regula- 72]. A gene knock-out study conducted by Jian et al. [73] tion of cytokine secretion involved in immune response revealed that C6KTD2 is essential particularly during (GO:0002740), and positive regulation of interferon- the blood stage of the parasite, thus targeting it in drug gamma production (GO:0032729) serves as immu- research is important. Interactome analysis on the host nological mediating processes that influence disease functional network revealed (P22301 (IL10), P05362 susceptibility by either conferring protection or influenc- (ICAM1), P01375 (TNF), P30480 (HLA-B), P16284 ing disease progress. Activation and regulation of NLRP3 (PECAM1), O00206 (TLR4)) as key targets. These host inflammasomes, immune system receptors, controls candidate key proteins are involved in immune response the activation of caspase-1 and induce inflammation in and resistance against malaria infection including severe response to infectious pathogens [81]. Due to their influ- and cerebral malaria, thus, critical targets for adjunctive ence on a wide range of diseases, their dysfunction results and antibody-based host-directed therapy for malaria in the initiation or progression of diseases. Endothelial control [74–76]. Importantly, studies have shown the cell apoptosis has been studied to contribute to malaria need to complement artemisinin derivatives with host- severity. For instance, haem-induced microvasculature directed therapy involved in immune modulation to help endothelial cell apoptosis mediated by proinflammatory effectively control and treat severe malaria and cerebral and proapoptotic pathways contributes significantly to malaria [77]. This may contribute significantly to improve severe malaria. treatment efficacy, reduce disease-associated complex- In addition, the pathways of immune tolerance and ity, reduce malaria-associated mortality and morbidity potential resistance development among the host and as well as slow artemisinin resistance development. In pathogen key targets were investigated by analysing the both the parasite and host-parasite functional network, shortest paths between these genes within the host–P. the functional interactions between hubs formed by falciparum functional network. The results showed that C6KTD2 and C6KTB7 were identified (Fig.  2B). This these shortest paths between the candidate genes or pro- finding suggests the functional relatedness of these pro- teins are mediated by host genes involved in cell regula- teins and their modularity within the parasite to jointly tory activities and general cell integrity. regulate post-translational modification processes. Hav- Shortest path analysis further revealed human ing established that nodes within a cluster might be immune-related genes and pathways that could be over- involved in the same biological process, it is, therefore, whelmed by the pathogen, knowing that the pathol- possible that these key proteins within the clusters con- ogy of malaria is immune-mediated and inflammatory tribute significantly to similar processes [78]. response-driven. Such inhibition could result in reduced 23 significantly enriched malaria-related biological anti-inflammatory responses thus limiting the produc- processes described in (Table  3) were identified. These tion and possible cytopathic effects of cytokines [82]. gene ontology groups comprised of those involved in The analysis revealed potential pathways between host cell immune and inflammatory responses, regulation malaria-associated candidate key protein O00206 (Toll- and production of transcription factors, biosynthetic like receptor 4, TLR4) and pathogen proteins C6KTB7 processes, cell–cell adhesion, cell signalling, and cell (Putative E3 ubiquitin-protein ligase protein PFF1365c) A gamah et al. Malar J (2021) 20:421 Page 15 of 20 Fig. 4 A Functional interactions between C6KTD2 and C6KTB7 subnetwork in the unified host–pathogen functional network. The shared host proteins (yellow nodes) are involved in protein ubiquitination, positive regulation of cell apoptotic process, signal transduction, regulatory processes, and histone methylation. B Predicted shortest path network that could influence resistance and parasite adaptiveness between C6KTB7 (green node) and O00206 (bottom sky blue node) via co–targets (central sky blue nodes) in the host–pathogen network. C Predicted shortest path network that could influence resistance and parasite adaptiveness between C6KTD2 (green node) and O00206 (bottom sky blue node) via mediators (central sky blue nodes) in the host–pathogen network and C6KTD2 (Putative histone-lysine N-methyltrans- of pro-inflammatory mediators, such as TNF and nitric ferase 1, SET1) that could account for unrestrained par- oxide [50, 83]. It also induces the expression of adhesion asite growth and severe complications. Experimental molecules on endothelial cells [50]. This may suggest that findings have revealed that activation of TLRs induces PECAM1, ICAM1, and TNF are from the downstream the production of nitric oxide and synthesis of pro- signalling cascade generated by TLR4 [83]. inflammatory cytokines, such as TNF and IL-1β [50, 83]. Severe malaria is associated with an increased level of Of note, activation of TLR4 induces macrophage release pro-inflammatory cytokines (T helper 1 (Th1) cytokines) Agamah et al. Malar J (2021) 20:421 Page 16 of 20 Fig. 4 continued such as interleukin (IL)-12, IL-8, and interferon (IFN)-γ ranked based on the enriched similarity score. The results in the affected person which helps to modulate defence revealed certolizumab pegol and golimumab as hits for against the infection and limit disease progression [59, the monoclonal antibody category, pomalidomide for 82]. This is attributed to the fact that the severity of the immunomodulator category, ginseng for the herbs malaria is proportional to the flawlessness in the host and natural product category, ruxolitinib for the Janus inflammatory response. kinase inhibitors, anistreplase for the thrombolytic agent TLR4, a pathogen-recognition receptor, detects path- category, and chloroquine for the anti-malarial category. ogen-associated molecular mechanisms in the body and Additional file  9: Table  S6 describes the known activity initiates immune response through activation of signal- and the original therapeutic purpose of the potentially ling cascades such as nuclear factorkB, mitogen-activated repurposable drugs identified. protein kinase (MAPK), and Plasmodium antigens [59]. TLR4 and its immune-related signalling pathways have Conclusions been reported to contribute significantly to P. falciparum With the gradual emergence and spread of malaria drug growth and malaria pathogenesis, such that dysregula- resistance, considering other potential drug targets and tion and dysfunction of the gene increase malaria sever- drug candidates are essential to increase the longev- ity, symptomatic malaria, severe malaria anaemia, and ity of existing drugs as well as develop alternative treat- resistance in Africa [84]. This suggests that deleterious ment options. In this research, integrative computational activation of TLR4 by C6KTB7 and C6KTD2 will signifi- methods were leveraged to (1) predict potential drug cantly contribute to parasite survival and disease suscep- targets for both human host and pathogen-directed drug tibility thereby causing severe pathological conditions. discovery, (2) predict drug candidates that could be re- Finally, a semantic similarity approach was imple- engineered for malaria treatment and, (3) identify biolog- mented to identify 113 diseases like malaria (Additional ical processes and pathways that could be overwhelmed file  8: Table  S5) that facilitated the prediction of 26 by the pathogen to increase within-host survival. potential repurposable drug hits, spanning across anti- The analysis revealed that repurposable drugs involved malarials, monoclonal antibodies, immunomodulators, in regulating host immune response to inflammatory- herbs, natural products, Janus kinase inhibitors, and driven disorders and/or inhibiting residual malaria infec- thrombolytic agents, that can be computationally and tion may enable appropriate malaria treatment. Of note, experimentally modified for parasite or host-directed the potential to treat malaria using inhibitors or drugs malaria treatment. Drug hits for each category were that target the proteasome component and/or proteins A gamah et al. Malar J (2021) 20:421 Page 17 of 20 involved in the parasite’s post-translational modification network. Figures A and D show that the majority of nodes are character- such as C6KTB7 and C6KTD2 have been established. ized by a relatively high betweenness and degree score. This depicts the However, exploring these targets for drug and vaccine small-world property of the network whereby non-neighboring nodes within the network can interact through influential nodes. Figures B and development is yet to be fully achieved. Both C6KTD2 C show that lower degree nodes are usually in close interaction thus, and C6KTB7 proteins have no crystallized structure yet, suggesting that such nodes are involved in similar processes or pathways, but the availability of other homologs could be explored thus execute the function within a smaller compartment (low-level using homology modelling approach to model the pro modularity) of the system, and the effect is transmitted by central nodes - with relatively higher degree and betweenness. Figures C and F suggest teins. The generated homology models could be the start- that signalling (flow of information) within the biological system is highly ing point for novel drug discovery and structure-based influenced by nodes with relatively high betweenness. Such nodes are characterized by relatively high degree and closeness and are known to studies to identify potential inhibitors. Additionally, the transmit signals generated as a result of low-level modularity between host protein targets predicted have solved structures that nodes. can be harnessed for structure-based drug discovery to identify potential inhibitors for malaria research. In summary, the uniqueness of the integrative network Additional file 2: Figure S2A. Summary results for parasite network framework lies in the input datasets, scoring metrics/ clustering. schemes, clustering algorithm, and the criteria defined for the various analysis which translates into the findings Additional file 3: Figure S2B. Summary results for host network from this study. The integrative network-based approach clustering. incorporates interologs, sequence blast interactions, and Additional file 4: Table S1. Description of various datasets and databases protein–protein interaction data from the literature, as used for the study. well as the STRING, IntAct, MINT, and BIOGRID data- Additional file 5: Table S2. Malaria-associated genes were retrieved by mapping significant SNPs to the gene level. The table entails the gene’s bases. In addition, the network approach implements functional network centrality scores, including betweenness, degree, and a scalable hierarchical agglomerative clustering model, closeness. based on modularity optimization, to cluster the net- Additional file 6: Table S3. Degree, closeness, and betweenness central- work into communities by leveraging candidate genes. ity score of C6KTD2 and C6KTB7 within the parasite unified functional This is then followed by network topology analysis to network. evaluate the topological features (degree, betweenness, Additional file 7: Table S4. Degree, closeness, and betweenness central- ity score for host candidate key proteins within the human functional and closeness) of the malaria candidate genes to identify network. hubs genes/proteins. The semantic similarity measures Additional file 8: Table S5. Predicted malaria–similar diseases identified implemented coupled with literature evidence helped to using semantic similarity approach. ESS represents the estimated enriched identify diseases similar to malaria and potential repur- similarity scores. posable drug candidates. Additional file 9: Table S6. Predicted repurposable drug hits identified Like other computational approaches which need using semantic similarity approach. validation through further functional study, our find- Additional file 10. Supplementary method. ings presented can inform functional study for potential experimental and clinical validation. Extended computa- Acknowledgements tional analysis of this work would consider incorporating We acknowledge the staff, colleagues from the Division of Human Genetics and H3Africa Coordinating Center, University of Cape Town. We acknowl- non-reviewed protein data, other omics level datasets, edge members of the Trusted World of Corona (TWOC) Consortium. We also and drug-drug interaction information. acknowledge the staff and colleagues from the Center for Molecular and Bio- molecular Informatics (CMBI), Radboud University Medical Center, Nijmegen. Computations were performed using facilities provided by the Centre for High-Performance Computing (https://u sers.c hpc. ac.z a/, South Africa). Abbreviations WHO: World Health Organization; SNP: Single Nucleotide Polymorphism; Authors’ contributions GWAS: Genome-wide Association Study; ESS: Enriched Similarity Score; PPIN: EC and GM designed the study, FA performed the data analysis and drafted Protein–Protein Interaction Network; ACT : Artemisinin-based Combination the manuscript. EC, DD, MS, AG, GM contributed to the data analysis and Therapy. revision of the manuscript and supervised the work. All authors read and approved the final manuscript. Supplementary Information Authors’ information The online version contains supplementary material available at https://d oi. Francis E. Agamah, PhD student in Human Genetics at the division of Human org/ 10. 1186/s 12936-0 21-0 3955-0. Genetics, Department of Pathology University of Cape Town. Email: agm- fra001@myuct.ac.za/francisagamahh@gmail.com. Additional file 1: Figure S1. Relationship between the degree, between- Delesa Damena, PhD in Human Genetics at the division of Human Genetics, ness, and closeness centrality measures in the host-parasite assembled Department of Pathology University of Cape Town. Email: delesa_damenaa@ functional network. Figures A, B and C show the relationship observed yahoo.com /damenadelesa@gmail.com. in the parasite network whereas Figures D, E, and F represent the host Agamah et al. Malar J (2021) 20:421 Page 18 of 20 Michelle Skelton, PhD in Human Genetics at, Computational Biology Division, Department of Integrative Biomedical Sciences, University of Cape Town. References Email: michelle.skelton@uct.ac.za. 1. WHO. World malaria report 2019. Geneva, World Health Organization, Anita Ghansah, PhD in Genetic Epidemiology at the London School of 2019. Hygiene and Tropical Medicine. Senior Researcher at Noguchi Memorial 2. Takala-Harrison S, Laufer MK. Antimalarial drug resistance in Africa: key Institute for Medical Research, University of Ghana. Email: aghansah@noguchi. lessons for the future. Ann N Y Acad Sci. 2015;1342:62–7. ug.edu.gh. 3. Amor A, Toro C, Fernandez-Martinez A, Baquero M, Benito A, Berzosa P. Gaston K. Mazandu, PhD in Bioinformatics, Senior Lecturer at the Division of Molecular markers in Plasmodium falciparum linked to resistance to anti- Human Genetics, Department of Pathology, University of Cape Town. Email: malarial drugs in samples imported from Africa over an eight-year period gmazandu@gmail.com. (2002–2010): impact of the introduction of artemisinin combination Emile R. Chimusa, PhD in Bioinformatics. Associate Professor at the Division of therapy. Malar J. 2012;11:100. Human Genetics, Department of Pathology, University of Cape Town. Email: 4. Ouji M, Augereau J-M, Paloque L, Benoit-Vical F. Plasmodium falciparum emile.chimusa@uct.ac.za. resistance to artemisinin-based combination therapies: a sword of Damo- cles in the path toward malaria elimination. Parasite. 2018;25:24. Funding 5. Miraclin TA, Matthew A, Rupali P. Decreased response to artemisinin com- This work was supported through the DELTAS Africa Initiative [DELGEME bination therapy in falciparum malaria: a preliminary report from South grant 107740/Z/15/Z]. The DELTAS Africa Initiative is an independent funding India. Trop Parasitol. 2016;6:85–6. scheme of the African Academy of Sciences (AAS)’s Alliance for Accelerating 6. Antony HA, Parija SC. Antimalarial drug resistance: an overview. Trop Excellence in Science in Africa (AESA) and supported by the New Partnership Parasitol. 2016;6:30–41. for Africa’s Development Planning and Coordinating Agency (NEPAD Agency) 7. Uwimana A, Umulisa N, Venkatesan M, Svigel SS, Zhou Z, Munyaneza T, with funding from the Wellcome Trust [DELGEME grant 107740/Z/15/Z] and et al. Association of Plasmodium falciparum kelch13 R561H genotypes the UK government. Also, this work was supported through the University of with delayed parasite clearance in Rwanda: an open-label, single-arm, Cape Town, internal funding, and the National Research Foundation of South multicentre, therapeutic efficacy study. Lancet Infect Dis. 2021;21:1120–8. Africa for funding (NRF) [grant # RA171111285157/119056]. This work was 8. Balikagala B, Fukuda N, Ikeda M, Katuro OT, Tachibana SI, Yamauchi M, partially funded by an LSH HealthHolland grant to the TWOC consortium, et al. Evidence of artemisinin-resistant malaria in Africa. N Engl J Med. a large-scale infrastructure grant from the Dutch Organization of Scientific 2021;385:1163–71. Research (NWO) to the Netherlands X-omics initiative (184.034.019), and a 9. Lubell Y, Dondorp A, Guérin P, Drake T, Meek S, Ashley E, et al. Artemisinin Horizon2020 research grant from the European Union to the EATRIS-Plus resistance–modelling the potential human and economic costs. Malar J. infrastructure project (grant agreement: No 871096). Some of the authors are 2014;13:452. supported in part by the National Institutes of Health (NIH) Common Fund 10. Conrad MD, Rosenthal PJ. Antimalarial drug resistance in Africa: the calm under grant numbers 1U2RTW012131-01 (COBIP), U24HG006941 (H3ABioNet) before the storm? Lancet Infect Dis. 2019;19:e338–51. and 1U01HG007459-01 (SADaCC). The content of this publication is solely the 11. White NJ. Antimalarial drug resistance. J Clin Invest. 2004;113:1084–92. responsibility of the authors and does not necessarily represent the official 12. Gatton ML, Martin LB, Cheng Q. Evolution of resistance to sulfadoxine- views of the funders. pyrimethamine in Plasmodium falciparum. Antimicrob Agent Chemother. 2004;48:2116–23. Availability of data and materials 13. Fairhurst RM, Dondorp AM. Artemisinin-resistant Plasmodium falciparum All the scripts and data used in this manuscript are available at https:// github. malaria. Microbiol Spectr. 2016;4(10):1128. com/ franc is- agamah/N etwo rk_ Analys is_ Malari a. Online Mendelian Inherit- 14. Acharya P, Garg M, Kumar P, Munjal A, Raja KD. Host–parasite interactions ance in Man (http:// www. omim.o rg). Supplementary data and figures are in human malaria: clinical implications of basic research. Front Microbiol. available online. 2017;8:889. 15. Clayton AM, Dong Y, Dimopoulos G. The Anopheles innate immune system in the defense against malaria infection. J Innate Immun. Declarations 2014;6:169–81. 16. Luckhart S, Pakpour N, Giulivi C. Host–pathogen interactions in malaria: Ethics approval and consent to participate cross-kingdom signaling and mitochondrial regulation. Curr Opin Immu- Not applicable. nol. 2015;36:73–9. 17. Su XZ, Zhang C, Joy DA. Host-malaria parasite interactions and impacts Consent for publication on mutual evolution. Front Cell Infect Microbiol. 2020;10:587933. Not applicable. 18. Ramaprasad A, Pain A, Ravasi T. Defining the protein interaction network of human malaria parasite Plasmodium falciparum. Genomics. Competing Interests 2012;99:69–75. The authors declare that the research was conducted in the absence of any 19. Agamah FE, Mazandu GK, Hassan R, Bope CD, Thomford NE, Ghansah A, commercial or financial relationships that could be construed as a potential et al. Computational/in silico methods in drug target and lead prediction. conflict of interest. Therefore, the authors declare that they have no compet- Brief Bioinform. 2020;21:1663–75. ing interests. 20. Zuck M, Austin LS, Danziger SA, Aitchison JD, Kaushansky A. The promise of systems biology approaches for revealing host pathogen interactions Author details in malaria. Front Microbiol. 2017;8:2183. 1 Division of Human Genetics, Department of Pathology, Institute of Infec- 21. Network MGE. Insights into malaria susceptibility using genome-wide tious Disease and Molecular Medicine, Faculty of Health Sciences, University data on 17,000 individuals from Africa. Asia and Oceania Nat Commun. of Cape Town, Cape Town, South Africa. 2 Computational Biology Division, 2019;10:5732. Department of Integrative Biomedical Sciences, Institute of Infectious Disease 22. Smigielski EM, Sirokin K, Ward M, Sherry ST. dbSNP: a database of single and Molecular Medicine, Faculty of Health Sciences, University of Cape Town, nucleotide polymorphisms. Nucleic Acids Res. 2000;28:352–5. Cape Town, South Africa. 3 College of Health Sciences, Noguchi Memorial Insti- 23. Sherry ST, Ward M, Kholodov M, Baker J, Phan L, Smigielski EM, et al. tute for Medical Research, University of Ghana, P.O. Box LG 581, Legon, Ghana. dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 4 African Institute for Mathematical Sciences, 5-7 Melrose Road, Muizenberg, 2001;29:308–11. Cape Town 7945, South Africa. 24. UniProt Consortium. The Universal Protein Resource (UniProt) in 2010. Nucleic Acids Res. 2010;38(Database issue):D142–8. Received: 30 July 2021 Accepted: 16 October 2021 A gamah et al. Malar J (2021) 20:421 Page 19 of 20 25. Mazandu GK, Mulder NJ. Scoring protein relationships in functional inter- proteins binding ICAM-1 and is a target of cross-reactive, adhesion-inhibi- action networks predicted from sequence data. PLoS One. 2011;6:e18607. tory antibodies. J Immunol. 2013;190:240–9. 26. Hunter S, Apweiler R, Attwood TK, Bairoch A, Bateman A, Binns D, et al. 52. Wishart DS, Knox C, Guo AC, Shrivastava S, Hassanali M, Stothard P, et al. InterPro: the integrative protein signature database. Nucleic Acids Res. DrugBank: a comprehensive resource for in silico drug discovery and 2009;37(Database issue):D211–5. exploration. Nucleic Acids Res. 2006;34(Database issue):D668–72. 27. Croft D, O’Kelly G, Wu G, Haw R, Gillespie M, Matthews L, et al. Reactome: 53. Fox CS. Using human genetics to drive drug discovery: a perspective. Am a database of reactions, pathways and biological processes. Nucleic Acids J Kidney Dis. 2019;74:111–9. Res. 2011;39(Database issue):D691–7. 54. Chen Y, Xu R. Network-based gene prediction for Plasmodium falcipa- 28. Kerrien S, Aranda B, Breuza L, Bridge A, Broaches-Carter F, Chen C, et al. rum malaria towards genetics-based drug discovery. BMC Genomics. The IntAct molecular interaction database in 2012. Nucleic Acids Res. 2015;16(Suppl 7):S9. 2012;40(Database issue):D841-6. 55. Hua S. Targeting sites of inflammation: intercellular adhesion molecule-1 29. Licata L, Briganti L, Peluso D, Perfetto L, Iannucelli M, Galeota E, et al. as a target for novel inflammatory therapies. Front Pharmacol. 2013;4:127. MINT, the molecular interaction database: 2012 update. Nucleic Acids 56. Rives AW, Galitski T. Modular organization of cellular networks. Proc Natl Res. 2012;40(Database issue):D857–61. Acad Sci USA. 2003;100:1128–33. 30. Chatr-Aryamontri A, Oughtred R, Boucher L, Rust J, Chang C, Kolas NK, 57. Chen LC, Yeh HY, Yeh CY, Arias CR, Soo VW. Identifying co-targets to fight et al. The BioGRID interaction database: 2017 update. Nucleic Acids Res. drug resistance based on a random walk model. BMC Syst Biol. 2012;6:5. 2017;45(Database issue):D369–79. 58. Belachew EB. Immune response and evasion mechanisms of Plasmodium 31. Wuchty S, Ipsaro JJ. A draft of protein interactions in the malaria parasite falciparum parasites. J Immunol Res. 2018;2018:6529681. P. falciparum. J Proteome Res. 2007;6:1461–70. 59. Gowda D, Wu X. Parasite recognition and signaling mechanisms in innate 32. Wuchty S. Topology and weights in a protein domain interaction immune responses to malaria. Front Immunol. 2018;9:3006. network–a novel way to predict protein interactions. BMC Genomics. 60. Clark IA, Alleva LM, Mills AC, Cowden WB. Pathogenesis of malaria and 2006;7:122. clinically similar conditions. Clin Microbiol Rev. 2004;17:509–39. 33. Wuchty S. Rich-club phenomenon in the interactome of P. falciparum- 61. Murphy SC, Breman JG. Gaps in the childhood malaria burden in Africa: -artifact or signature of a parasitic life style? PLoS One. 2007;2:e335. cerebral malaria, neurological sequelae, anemia, respiratory distress, 34. Wuchty S, Adams JH, Ferdig MT. A comprehensive Plasmodium falciparum hypoglycemia, and complications of pregnancy. Am J Trop Med Hyg. protein interaction map reveals a distinct architecture of a core interac- 2001;64(1_suppl):57–67. tome. Proteomics. 2009;9:1841–9. 62. Briglia M, Fazio A, Faggio C, Laufer S, Alzoubi K, Lang F. Triggering 35. LaCount DJ, Vignali M, Chettier R, Phansalkar A, Bell R, Hesselberth JR, of suicidal erythrocyte death by ruxolitinib. Cell Physiol Biochem. et al. A protein interaction network of the malaria parasite Plasmodium 2015;37:768–78. falciparum. Nature. 2005;438:103–7. 63. Francischetti IM, Seydel KB, Monteiro RQ. Blood coagulation, inflamma- 36. Bossi A, Lehner B. Tissue specificity and the human protein interaction tion, and malaria. Microcirculation. 2008;15:81–107. network. Mol Syst Biol. 2009;5:260. 64. Apweiler R, Bairoch A, Wu CH, Barker WC, Boeckmann B, Ferro S, et al., 37. Mazandu GK, Mulder NJ. Generation and analysis of large-scale data- UniProt: the Universal Protein knowledgebase. Nucleic Acids Res. driven Mycobacterium tuberculosis functional networks for drug target 2004;32(Database issue):D115–9. identification. Adv Bioinformatics. 2011;2011:801478. 65. Ponts N, Yang J, Chung DK, Prudhomme J, Girke T, Horrocks P, et al. 38. Mulder NJ, Akinola RO, Mazandu GK, Rapanoel H. Using biological Deciphering the ubiquitin-mediated pathway in apicomplexan parasites: networks to improve our understanding of infectious diseases. Comput a potential strategy to interfere with parasite virulence. PLoS One. Struct Biotechnol J. 2014;11:1–10. 2008;3:e2386. 39. Blondel VD, Guillaume JL, Lambiotte R, Lefebvre E. Fast unfolding of com- 66. Hamilton MJ, Lee M, Le Roch KG. The ubiquitin system: an essential com- munities in large networks. J Stat Mech. 2008;10:P10008. ponent to unlocking the secrets of malaria parasite biology. Mol Biosyst. 40. Emmons S, Kobourov S, Gallant M, Börner K. Analysis of network 2014;10:715–23. clustering algorithms and cluster quality metrics at scale. PLoS One. 67. Villard V, Agak GW, Frank G, Jafarshad A, Servis C, Nébié I, et al. Rapid 2016;11:e0159161. identification of malaria vaccine candidates based on alpha-helical coiled 41. Waltman L, Van Eck NJ. A smart local moving algorithm for large-scale coil protein motif. PLoS One. 2007;2:e645. modularity-based community detection. Eur Phys J B. 2013;86:471. 68. Aminake MN, Arndt HD, Pradel G. The proteasome of malaria parasites: a 42. Rosvall M, Bergstrom CT. Maps of random walks on complex networks multi-stage drug target for chemotherapeutic intervention? Int J Parasitol reveal community structure. Proc Natl Acad Sci USA. 2008;105:1118–23. Drugs Drug Resist. 2012;2:1–10. 43. Raghavan UN, Albert R, Kumara S. Near linear time algorithm to 69. Sharma M, Dhiman C, Dangi P, Singh S. Designing synthetic drugs detect community structures in large-scale networks. Phys Rev E Stat. against Plasmodium falciparum: a computational study of histone-lysine 2007;76:036106. N-methyltransferase (PfHKMT). Syst Synth Biol. 2014;8:155–60. 44. Harris MA, Clark J, Ireland A, Lomax J, Ashburner M, Foulger R, et al. The 70. Doug Chung D-W, Le Roch KG. Targeting the Plasmodium ubiquitin/pro- Gene Ontology (GO) database and informatics resource. Nucleaic Acids teasome system with anti-malarial compounds: promises for the future. Res. 2004;32(Database issue):D258–61. Infect Disord Drug Targets. 2010;10:158–64. 45. Aurrecoechea C, Brestelli J, Brunck BP, Dommer J, Fischer S, Garija B, et al. 71. Cui L, Fan Q, Cui L, Miao J. Histone lysine methyltransferases and dem- PlasmoDB: a functional genomic database for malaria parasites. Nucleic ethylases in Plasmodium falciparum. Int J Parasitol. 2008;38:1083–97. Acids Res. 2009;37(Database issue):D539–43. 72. Kaur I, Zeeshan M, Saini E, Kaushik A, Mohmmed A, Gupta D, et al. 46. Aoki KF, Kanehisa M. Using the KEGG database resource. Curr Protoc Widespread occurrence of lysine methylation in Plasmodium falciparum Bioinformatics. 2005;Chapt 1:Unit 1.12. proteins at asexual blood stages. Sci Rep. 2016;6:35432. 47. McDonald JH. Handbook of biological statistics. Vol. 2. 2009: Sparky 73. Jiang L, Mu J, Zhang Q, Ni T, Srinivasan P, Ryavara K, et al. PfSETvs House Publishing, Baltimore, MD. methylation of histone H3K36 represses virulence genes in Plasmodium 48. Kibbe, W.A., et al., Disease Ontology 2015 update: an expanded and falciparum. Nature. 2013;499:223–7. updated database of human diseases for linking biomedical knowl- 74. Dunst J, Kamena F, Matuschewski K. Cytokines and chemokines in cer- edge through disease data. Nucleic Acids Res. 2015;43((Database ebral malaria pathogenesis. Front Cell Microbiol. 2017;7:324. issue):D1071–8. 75. Kumar R, Ng S, Engwerda C. The role of IL-10 in malaria: a double edged 49. Mathur S, Dinakarpandian D. Finding disease similarity based on implicit sword. Front Immunol. 2019;10:229. semantic similarity. J Biomed Inform. 2012;45:363–71. 76. Franklin BS, Ishizaka ST, Lamphier M, Gusovsky F, Hansen H, Rose J, 50. Gazzinelli RT, Kalantari P, Fitzgerald KS, Golenbock DT. Innate sensing of et al. Therapeutical targeting of nucleic acid-sensing Toll-like recep- malaria parasites. Nat Rev Immunol. 2014;14:744–57. tors prevents experimental cerebral malaria. Proc Natl Acad Sci USA. 51. Bengtsson A, Joergensen L, Rask TS, Olsen RW, Andersen MA, Turner L, 2011;108:3689–94. et al. A novel domain cassette identifies Plasmodium falciparum PfEMP1 Agamah et al. Malar J (2021) 20:421 Page 20 of 20 77. Varo R, Crowley VM, Sitoe A, Madrid L, Serguides L, Kain KC, et al. Adjunc- 83. Krishnegowda G, Hajjar AM, Zhu J, Douglass EJ, Uematsu S, Akira S, tive therapy for severe malaria: a review and critical appraisal. Malar J. et al. Induction of proinflammatory responses in macrophages by the 2018;17:47. glycosylphosphatidylinositols of Plasmodium falciparum: cell signaling 78. Mazandu GK, Chimusa ER, Rutherford K, Zekeng EG, Gebremariam ZZ, receptors, glycosylphosphatidylinositol (GPI) structural requirement, and Onifade MY, et al. Large-scale data-driven integrative framework for regulation of GPI activity. J Biol Chem. 2005;280:8606–16. extracting essential targets and processes from disease-associated gene 84. Greene JA, Moormann AM, Vulule J, Bocharie MJ, Zimmerman PA, Kazura data sets. Brief Bioinform. 2018;19:1141–52. JW. Toll-like receptor polymorphisms in malaria-endemic populations. 79. Tripathi AK, Sha W, Shulaev V, Stins MF, Sullivan DJ. Plasmodium falcipa- Malar J. 2009;8:50. rum-infected erythrocytes induce NF-kappa B regulated inflammatory 85. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment pathways in human cerebral endothelium. Blood. 2009;114:4243–52. search tool. J Mol Biol. 1990;215:403–10. 80. Lyke KE, Fernández-Vina MS, Cao K, Hollenbach J, Coulibaly D, Kone AK, 86. von Mering C, Huynen M, Jaeggi D, Schmidt S, Bork P, Snel B. STRING: a et al. Association of HLA alleles with Plasmodium falciparum severity in database of predicted functional associations between proteins. Nucleic Malian children. Tissue Antigens. 2011;77:562–71. Acids Res. 2003;31:258–61. 81. Guo HT, Callaway JB, Ting JP. Inflammasomes: mechanism of action, role in disease, and therapeutics. Nat Med. 2015;21:677–87. 82. Oyegue-Liabagui SL, Bouopda-Tuedom AG, Kouna LC, Maghendji-Nzo- Publisher’s Note ndo S, Nzoughe H, Tchitoula-Makaya N, et al. Pro- and anti-inflammatory Springer Nature remains neutral with regard to jurisdictional claims in pub- cytokines in children with malaria in Franceville. Gabon Am J Clin Exp lished maps and institutional affiliations. Immunol. 2017;6:9–20. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced rese archers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions