Additional file 1: of PPI network analyses of human WD40 protein family systematically reveal their tendency to assemble complexes and facilitate the complex predictions

Figure S1. Workflow of this study. Figure S2. Overview of the ALL-PPI network. Figure S3. Degree distributions of nodes in two networks. Figure S4. Percentage of WD40 proteins in each k-core subnetwork during the decomposition of ALL-PPI network. Figure S5. Distributions of average PCCs of WD40 hubs, non-WD40 hubs, and randomized data in ALL-PPI network. Figure S6. The number of reference complexes matched with predicted complexes obtained by different methods. Figure S7. The number of reference complexes matched with predicted complexes obtained from ALL-PPI network under different ω. Figure S8. Number of complexes in reference set matched with predicted complexes obtained from different PPI networks. Figure S9. k-core decomposition and localization of hubs in PPI network. Table S2. Counts of hubs in WD40 and non-WD40 proteins in ALL-PPI network. Table S3. Counts of hubs in WD40 and non-WD40 under different definitions. Table S4. Comparisons of centralities between WD40 and non-WD40 in two networks. Table S5. WD40 proteins in different layers obtained by k-core decomposition of ALL-PPI network. Table S6. Orthologs of MED16, GBLP, and CORO1C in model organisms. Table S7. Medians of average PCCs of WD40 hubs and non-WD40 hubs in two networks. Table S9. Statistics of complex predictions based on HC-PPI under different parameter settings. Table S11. Number of predicted complexes under different methods, and the matched numbers of reference complexes under different ω. Table S12. Comparisons of MMR for different prediction methods with ω> = 0.2. Table S13. Statistics of complex prediction obtained from ALL-PPI under different parameter settings. Table S14. Medians of co-expression scores for different complex sets under different expression dataset. (PDF 1014 kb