GENECFE-ANFIS: A NEURO-FUZZY INFERENCE SYSTEM TO INFER GENE-GENE INTERACTIONS BASED ON RECOGNITION OF MICROARRAY GENE EXPRESSION PATTERNS.

Chuang, Cheng-Long; Chen, Chung-Ming; Shieh, Grace S.; Jiang, Joe-Air

A neuro-fuzzy inference system that recognizes the expression patterns of genes in microarray gene expression (MGE) data, called GeneCFE-ANFIS, is proposed to infer gene interactions. In this study, three primary features are utilized to extract genes' expression patterns and used as inputs to the neuro-fuzzy inference system. The proposed algorithm learns expression patterns from the known genetic interactions, such as the interactions confirmed by qRT-PCR experiments or collected through text-mining technique by surveying previously published literatures, and then predicts other gene interactions according to the learned patterns. The proposed neuro-fuzzy inference system was applied to a public yeast MGE dataset. Two simulations were conducted and checked against 112 pairs of qRT-PCR confirmed gene interactions and 77 TFs (Transcriptional Factors) pairs collected from literature respectively to evaluate the performance of the proposed algorithm.

Biomedical Engineering: Applications, Basis & Communications, 2007, Vol 19, Issue 2, p71

1016-2372

Academic Journal

10.4015/S1016237207000112