Machine learning approach identifies prominent codons from different degenerate groups influencing gene expression in...

Area of Science:

• Microbiology
• Genomics
• Bioinformatics

Background:

• Codon usage bias (CUB) describes unequal synonymous codon usage.
• Differences in CUB are observed between high-expression genes (HEG) and low-expression genes (LEG).
• CUB patterns across diverse bacterial species remain incompletely understood.

Purpose of the Study:

• To identify codons with significant expression-dependent usage differences in bacteria.
• To investigate conserved CUB patterns across a wide range of bacterial species.
• To elucidate the role of CUB in gene expression regulation in bacteria.

Main Methods:

• Machine learning algorithms were employed to analyze codon usage in Escherichia coli.
• Comparative analysis of codon usage between HEG and LEG was performed.
• Codon usage patterns were examined across 683 bacterial species.

Main Results:

• In E. coli, Cys (UGU/UGC) and Lys (AAA/AAG) codons showed minimal expression influence.
• Ser (UCU), Leu (CUG), Gly (GGG), and Arg (CGG) codons were highly influenced by gene expression in E. coli.
• Across most bacterial species, Cys (UGU/UGC) and Ser (AGU/AGC) codons exhibited the least variation between HEG and LEG.
• Specific codons like CGA, CUG, GGG, GCC, ACC, AUA, and AUC were consistently influenced by gene expression across many species.

Conclusions:

• Codon usage bias plays a significant role in regulating gene expression across bacterial species.
• Certain codons demonstrate conserved behavior concerning gene expression levels in bacteria.
• This study highlights commonalities in CUB influencing gene expression across the bacterial domain.