论文部分内容阅读
Background: Codon usage bias (CUB), a balanced result of mutation and selection, is of great significance in better understanding gene evolution and providing insights on gene function and expression.However, quantitative comparative analysis of CUB in different human tissues remains unexplored, primarily limited by the comprehensiveness and accuracy of tissue expression data and the reliability of CUB estimation.Methods: Equipped with our newly-developed algorithm (Codon Deviation Coefficient) for accurate estimation of CUB, we performed a systematic investigation on CUB in 10 human tissues based on RNA-Seq data including testis, brain, adipose, breast, colon, heart, kidney, liver, lymph node and muscle.Results: We found that tissue-specific (TS) genes and housekeeping (HK) genes exhibit different CUB patterns.All TS genes across 10 tissues show relatively lower CUBs than HK genes.Among 10 tissues, brain-specific genes have lowest CUBs, whereas adipose-specific genes have highest CUBs.Considering the distribution of CUB, testis-specific genes show the broadest range of CUB, probably due to their large diversity of gene function.Contrastingly, liver-specific genes present the narrowest range of CUB, which may be related to their function exclusively responsible for metabolism.In addition, HK genes that are highly and consistently expressed in different tissues have a close correlation between CUB and gene expression level, whereas TS genes do not present such correlation.Conclusions: These results indicate that CUB is a fundamental mechanism in regulating gene expression in both prokaryotes and eukaryotes.Although there are multiple mechanisms complicating gene regulation in human, genes that perform basic functions may still be substantially regulated through CUB .