COMPARISONS OF CODING AND NONCODING SEQUENCES TOINFER THE ORIGIN OF CODON USAGE BIAS

Abstract

Codon usage bias is the preferential use of the subset of synonymous codons during translation. In this paper, the comparisons of normalized entropy and GC content between the sequence of coding regions of Escherichia coli k12 and noncoding regions (ncRNA, rRNA) of various organisms were done to shed light on the origin of the codon usage bias.The normalized entropy of the coding regions was found significantly higher than the noncoding regions, suggesting the role of the translation process in shaping codon usage bias. Further, when the position specific GC content of both coding and noncoding regions was analyzed, the GC2 content in coding regions was lower than GC1 and GC2 while in noncoding regions, the GC1, GC2, GC3 contents were approximately equal. This discrepancy is explained by the biased mutation coupled with the presence and absence of selection pressure. The accumulation of CG content occurs in the sequences due to mutation bias in DNA repair and recombination process. In noncoding regions, the mutation is harmful and thus, selected against while due to the degeneracy of codons in coding regions, a mutation in GC3 is neutral and hence, not selected. Thus, the accumulation of GC content occurs in coding regions, and thus codon usage bias occurs.