Abstract

The demand for extracting sophisticated features, capable of effectively predicting gene interaction networks, from DNA or RNA sequences has increased in computational biology. The epigenetic modifications along with their patterns have been intensely recognized as appealing features affecting on gene interaction. However, studying sequenced-based features highly correlated to this key element has remained limited. In this paper, classification of 23 genes in PPAR signaling pathway associated with muscle fat tissue in human was proposed based on statistical distributions of the specified RNA5-mers abundance. Then, we suggested that these 5- mers highly correlated to epigenetic modifications can efficiently categorize the different gene interactions, particularly co-expression interaction and physical interaction. Our results were evaluated according to GeneMania web interface and shows that the geometric distribution of 5 mers in the epigenetic modifications region indicates the proportion of most physical interactions and the Poisson distribution the proportion of most Co-expression between genes