Fig. 1

Network module analysis of protein sequences from six Plasmodium species. (A) Workflow of network analysis. Sequence alignment was performed using phmmer, which resulted in a matrix representing sequence similarity. The matrix was converted into a large, disconnected network. Graph shortest path was introduced to identify all the components of the network and the BGLL algorithm was adopted to discover the modules within each component by DFS strategy. The number of proteins from six Plasmodium species in each module or cluster was calculated, which resulted in the enrichment of each cluster in six species. (B) Upper panel shows the number of network modules at different thresholds identified by BGLL. Arrow indicates the modularity chosen in this study. Lower panel shows the number of clusters at different enrichment. (C) Heatmap showing the clusters in six Plasmodium species. Hierarchical clustering analysis of six species based on the enrichment of 4693 clusters. Green, black, and red indicate cluster values equal to zero, one, and higher than one, respectively