Analysis of Codon Usage Bias in Salviae miltiorrhizae Based on Transcriptome Data

To reveal the characteristics of codon usage bias in the transcriptome of S. miltiorrhiza systematic analyses were performed on high-quality coding sequences (CDS) of S. miltiorrhiza with the aid of Codon W software and Perl scripts. The results showed that the overall guanine-cytosine (GC) content of the coding sequences in the S. miltiorrhiza transcriptome ranged from 25.74% to 71.69%, with an average of 48.61%. The GC content exhibited an unbalanced distribution across different codon positions: the mean GC content at the first codon position (GC₁) and the third codon position (GC₃) were close and significantly higher than that at the second codon position (GC₂). Based on the characteristics of neutrality plot and ENC plot, natural selection was confirmed to be the core driving force shaping the codon usage bias of S. miltiorrhiza. Correspondence analysis (COA) further clarified the regulatory role of genomic base composition in codon usage patterns. The codon adaptation index (CAI) of S. miltiorrhiza ranged from 0.073 to 0.525, and the effective number of codons (ENC) varied between 24.18 and 61.00, indicating a weak overall codon usage bias in the species. Through the analysis of relative synonymous codon usage (RSCU), a total of 18 optimal codons (e.g GCA, GCU, AGA) were identified, all of which ended with A or U bases except for UUG. This study clarified the characteristics of codon usage and the dominant driving factors in the S. miltiorrhiza transcriptome, thereby providing a theoretical basis for its germplasm identification, genetic engineering improvement, and optimization of heterologous expression vectors.