Abstract
Nucleotide compositional asymmetry (NCA) between leading and lagging strands (LeS and LaS) is dynamic and diverse among eubacterial genomes due to different mutation and selection forces. A thorough investigation is needed in order to study the relationship between nucleotide composition dynamics and gene distribution biases. Based on a collection of 364 eubacterial genomes that were grouped according to a DnaE-based scheme (DnaE1–DnaE1, DnaE2–DnaE1, and DnaE3–PolC), we investigated NCA and nucleotide composition gradients at three codon positions and found that there was universal G-enrichment on LeS among all groups. This was due to a strong selection for G-heading (codon position1 or cp1) codons and mutation pressure that led to more G-ending (cp3) codons. Moreover, a slight T-enrichment of LeS due to the mutation of cytosine deamination at cp3 was universal among DnaE1–DnaE1 and DnaE2–DnaE1 genomes, but was not clearly seen among DnaE3–PolC genomes, in which A-enrichment of LeS was proposed to be the effect of selections unique to
polC and a mutation bias toward A-richness at cp1 that may be a result of transcription-coupled DNA repair mechanisms. Furthermore, strand-biased gene distribution enhances the purine-richness of LeS for DnaE3–PolC genomes and T-richness of LeS for DnaE1–DnaE1 and DnaE2–dnaE1 genomes.