Supplementary MaterialsAdditional document 1 Clusters obtained using CASCADE for the yeast PPI network. 1. (a) cluster 1, size 411. (b) cluster 2, size 303. (c) cluster 3, size 240. (d) cluster 4, size 176. (e) cluster 5, size 170. (f) cluster 6, size 104. (g) cluster 7, size 96. (h) cluster 8, size 79. (i) cluster 9, size 78. (j) cluster 10, size 73. Each physique presents the percentile of proteins that are accordant with the top ten best accordant functional terms for each cluster. 1471-2105-9-64-S2.pdf (185K) GUID:?630FE655-3F06-4972-BB4D-7C17569DF098 Additional file 3 Normalized number of functional terms for each cluster detected by CACASDE. The first column is usually a cluster identifier; the Size column indicates the number of proteins in each cluster. The normalized numbers of functional terms in the MIPS functional hierarchy for each identified cluster are offered in the third, the fourth, and the fifth column. The number of functional conditions per each cluster is certainly normalized by its cluster size. The 3rd column symbolizes the normalized amount of functional conditions that are even more specific than 2nd level useful hierarchy. The 4th column represents the normalized amount of functional conditions that are even more specific than 3rd level useful hierarchy. The 5th column represents the normalized amount of functional conditions that are even more specific than 4th level useful hierarchy. 1471-2105-9-64-S3.pdf (65K) GUID:?C52BD956-76A7-4865-8C28-B63B933F0751 Extra file 4 Topological form of a cluster and its own useful annotations. Cluster 20 in Additional Document 1. (a) sub graph of Cluster 20 extracted from DIP PPI network. Each proteins is certainly annotated by MIPS useful category. (b) MIPS useful IDs and their corresponding Torisel cost literal brands. The very best accordant useful term is certainly boldfaced. 1471-2105-9-64-S4.pdf (65K) GUID:?58FBFAAF-2D46-40C6-91A9-3E2B8FA31967 Additional file 5 Topological form of a cluster and its own useful annotations. Cluster 21 in Additional Document 1. (a) sub graph of Cluster 21 extracted from DIP PPI network. Each proteins is certainly annotated by MIPS useful category. (b) MIPS useful IDs and their corresponding literal brands. The best accordant practical term is definitely FJX1 boldfaced. 1471-2105-9-64-S5.pdf (61K) GUID:?FFB16179-EA1D-4E53-A7B9-323419FFCCC0 Additional file 6 Topological shape of a cluster and its practical annotations. Cluster 22 in Additional File 1. (a) sub graph of Cluster 22 extracted from DIP PPI network. Each protein is definitely annotated by MIPS practical category. (b) MIPS practical IDs and their corresponding literal titles. The best accordant practical term is definitely boldfaced. 1471-2105-9-64-S6.pdf (60K) GUID:?E7DF73FD-8242-49AE-9E4A-7A4185E10F73 Additional file 7 Topological shape of a cluster and its practical annotations. Cluster 25 in Additional File 1. (a) sub graph of Cluster 25 extracted from DIP PPI network. Each protein is definitely annotated by MIPS practical category. (b) MIPS practical IDs Torisel cost and their corresponding literal titles. The Torisel cost best accordant practical term is definitely boldfaced. 1471-2105-9-64-S7.pdf (65K) GUID:?36C34AAA-B6DD-4579-BF2A-EC83B8B32B3A Abstract Background Quantitative characterization of the topological characteristics of protein-protein interaction (PPI) networks can enable Torisel cost the elucidation of biological practical modules. Here, we present a novel clustering methodology for PPI networks wherein the biological and topological influence of each protein on additional proteins is definitely modeled using the probability distribution that the series of interactions necessary to link a couple of distant proteins in the network happen within a time constant (the occurrence probability). Results CASCADE selects representative nodes for each cluster and iteratively refines clusters based on a combination of the occurrence probability and graph topology between every protein pair. The CASCADE approach is compared to nine competing methods. The clusters acquired by each technique are compared for enrichment of biological function. CASCADE generates larger clusters and the clusters recognized possess em p /em -values for biological function that are approximately 1000-fold better than the other methods on the yeast PPI network dataset. An important strength of CASCADE is definitely that the percentage of proteins that are discarded to produce clusters is much lower than the additional approaches which have an average discard rate of 45% on the yeast protein-protein interaction network. Summary CASCADE is effective at detecting biologically relevant clusters of interactions. Background Protein-protein interactions (PPI) and other.