Analysis of codon usage bias in the mitochondrial protein-coding genes of Oryza rufipogon

The mitochondrial genome of Oryza rufipogon Griff. was used to analyze the codon usage characteristics of protein-coding genes and the differences with Asian cultivated rice (O.sativa L.), and to explore the influencing factors of codon usage bias and codon evolution. Results showed that the effective number of codons (Nc) ranged from 45.32 to 61.00, indicating that codon bias was weak. GC content at the three codon positions was 49.18%, 42.67%, and 40.86%. The Nc value was significantly correlated with GC₃ but not with GC₁ or GC₂, suggesting that base composition at the third codon position had a greater effect on codon bias. In the corresponding analysis, the first axis showed 9.91% variation and was significantly correlated with GC_3s, Nc, CBI, and Fop. Furthermore, GC₁₂ showed non-significant correlation with GC₃. Codon bias in the mitogenome of O. rufipogon was mainly affected by natural selection. In addition, we identified 21 optimal codons, with most of the preferred synonymous codons ending in A or T. The mitochondrial codons showed convergent evolution with Oryza chloroplast codons, but different preferences with the nuclear genome. Based on neutrality plot analysis, PR2-plot analysis, and RSCU analysis, we found no significant differences among the three Oryza species. Our results also confirmed homogeneity in mitochondrial codon usage among the three Oryza species.