[Comparative studies on codon usage bias of Ganoderma lucidum based on analysis of genomic and transcriptomic data]

Codon usage bias is an important characteristic of genetic information transfer in organisms. Analysis of codon usage bias of different species is important for understanding the rules on genetic information transfer. The previous method for analysis of codon usage bias is mainly based on genomic data. However, this method is greatly limited, because the genome sequences of higher organisms are still not available up to now. In this study, we found that we could obtain the same optimal codons of Ganoderma lucidum (Curtis: Fr.) P. Karst based on its whole genomic data or large-scale transcriptomic data from its liquid-cultured hyphae, primordium and fruiting body, separately. This result indicated the feasibility to understand the codon usage bias based on the large-scale transcriptomic data. By calculating the proportion of rare codons of Escherichia coli and Saccharomyces cerevisiae in 26 terpene synthases (TS) of G. lucidum, we found that the rare codons of S. cerevisiae have a higher proportion in TS genes, while the rare codons of E. coli have relatively lower, suggesting that the TS genes of G. lucidum are possibly more difficult to be expressed in S. cerevisiae than in E. coli. Chemical synthesis of TS genes according to the yeast optimal codons will be an effective way to solve the problem on the mismatch of gene codon bias between the foreign genes and the host strain.