Application of Consensus Differential Network Analysis to Explore Host Response to Plasmodium cynomolgi infection

Background: Plasmodium cynomolgi infections of Macaca mulatta display similar clinical symptoms to Plasmodium vivax infections. Differences in the transcriptome between P. cynomolgi infected and healthy individuals have the potential to reveal underlying molecular mechanisms that characterize host response to P. vivax infection. Differential network analysis (DiNA) is a recent class of algorithms designed to find differences of network topology between states and these algorithms have been successfully applied in cancer research. Material and Methods: Here we propose a consensus based differential network analysis (cDiNA) to conduct comparative analysis of transcriptome data between four Plasmodium cynomolgi infected Macaca mulatta and five healthy controls. The analysis was carried out for bone marrow and whole blood transcriptome data separately. cDiNA overcomes the limitations of using a single metric to characterize topological differences between groups and complements existing transcriptome analysis result through the addition of differential topology analysis. Results and Conclusion: Of particular interest, our analysis reveals that the differential topology of the bone marrow transcriptome is enriched in oxidative phosphorylation and mitochondrial specific genes. Furthermore, the analysis of the whole blood transcriptome demonstrates enrichment of antigen processing and presentation, spliceosome and T-cell signaling pathway related genes. Our analysis provides novel insights into host transcriptomic responses to malaria infection, which can be used to gain a better understanding of the mechanistic causes underlying malaria pathology.