Codon Usage in Prokaryotes

Abstract

ABSTRACTAdvances in DNA sequencing led to accumulation of genomic data of a great number of prokaryotes belonging to different taxons. In this study we have analyzed 158 prokaryotic genomes for codon usage specificity. The latter is studied not only on the level of the whole genome (all genes) but also for three gene subsets: highly expressed genes, ribosomal protein genes and low expressed genes, using our own programs written in Visual Basic Studio (VBS), Visual Basic for Applications (VBA) and Perl. The local databases thus created were explored to compare the codon usage specificity of: a) different strains of one bacterial species; b) different subsets of genes in one species c) different bacterial genomes and d) to identify missing and atypical initiation codons. Finally, using correlation analysis we have also checked the correspondence between the formal bacterial taxonomy and Genome Theory, which is based on the codon usage pattern in bacteria. Our results revealed a great deviation in codon usage pattern between bacteria belonging to different taxons. They demonstrated also that this deviation was not randomly spread over the different codon groups (20 groups of synonymous codons). Comparing different subset of genes we show that the codon usage pattern of the low expressed genes is very similar to that of the all genes, whereas the patterns of the highly expressed, and particularly of the ribosomal protein genes, are highly distinguished. All these data are available at www.bio21.bas.bg/imb/codonusage and they can be used for optimization of codon composition of genes designed for heterologous expression in bacteria. The correlation analysis demonstrates that in many cases the formal bacterial taxonomy does not correspond to the Genome Theory.