The Classification of Chaos Game Representations of Human Genes and the Comparison of Chaos Game Representations between Human and Other Mammalian Homologous Genes

Abstract

This paper has found that there are four distinct patterns of chaos game representations (CGRs) of DNA sequences existing in human genes. Simultaneously it showed that the class I of CGRs seems to be related to highly evolutionary rate. Both classes II and III seem to be related to normal evolutionary rate, and only class IV is highly conservative which belings to slow evolutionary rate. As for the CGRs of oncogenes and antioncogenes, belong to different CGRs, thus it is impossible that they share a common ancestor (except ras family). In other words, this study does not support both the views of oncogene superfamily and of the characteristic pattern of CGRs of human oncogenes proposed by Jeffrey. We noted by comparison what the CGRs of human genes with ones of other mammalian homologous genes show much more similarity than the distincts pattern of CGRs in human genes. Therefore the paper highlighter that CGR is gene-specific other than genomic-specific, though the latter was proposed by Jeffrey and Hill et al. We consider that the CGR may be a new tool for molecular evolution study.