Identification of disrupted pathways in ulcerative colitis-related colorectal carcinoma by systematic tracking the dysregulated modules.

PURPOSE: The aim of this study was to identify the altered biological pathways associated with ulcerative colitis (UC)-related colorectal carcinoma (CRC) by systematic tracking the dysregulated modules from re-weighted protein- protein interaction (PPI) networks based on the expression profiles from normal, UC and various stages of CRC.

METHODS: We firstly recruited the UC- and CRC-related microarray data from ArrayExpress database, and obtained 8 expression profiles which contained 5 conditions (normal, UC, early stage CRC, stage II CRC and stage III CRC). Then, the PPI networks of normal and different disease stages were constructed and re-weighted using Pearson correlation coefficient (PCC). Next, the condition-specific modules were extracted from 5 PPI networks via clique-merging algorithm, and altered modules were captured on the basis of module correlation density (MCD). Subsequently, the gene compositions of altered modules and gene differential expressions in different disease stages were identified to screen the dysregulated genes. Finally, pathways enrichment analyses for the genes in altered modules and differentially expressed genes (DEGs) were implemented.

RESULTS: The extensive changes of gene correlations existed in 5 condition-specific PPI networks, which made different MCDs among different disease stages. The same number of modules (N=1952) were explored in 5 PPI networks. By comparing with normal condition, there were 463, 791, 1060 and 345 altered modules in UC, early stage CRC, stage II and III CRC, respectively. Overall, 77, 110, 170 and 110 common genes were identified between genes of altered modules and DEGs in UC, early stage CRC, stage II CRC and stage III CRC, respectively. Functional enrichment analyses indicated that cell cycle and oocyte meiosis were the common and most significant pathways in colonic diseases.

CONCLUSIONS: Tracking the altered modules from PPI networks is useful to uncover disrupted pathways in colonic diseases. Cell cycle and oocyte meiosis might be associated with the pathophysiological background of colonic diseases.