Genomic AT Bias Coupled with Amino Acid Metabolism Modulates Codon Usage.

Area of Science:

• Genomics
• Evolutionary Biology
• Bioinformatics

Background:

• Genome evolution results in specific codon and amino acid proportions.
• Genomic GC (guanine + cytosine) content influences these proportions.

Purpose of the Study:

• To develop a simplified maximum entropy model for predicting codon and amino acid abundances.
• To account for genomic GC content and metabolic costs in genome evolution.

Main Methods:

• A maximum entropy model was created, grouping codons by GC content and amino acid.
• The model incorporates codon cost based on genomic GC content and amino acid metabolic cost.
• Analysis was performed on over 50,000 genomes using seven interpretable parameters.

Main Results:

• Both codon and amino acid costs are crucial for accurate abundance predictions.
• The optimal model suggests a universal equilibrium genomic GC content below 50%.
• Amino acids were grouped based on Watson-Crick base pairing strength (GC vs. AU), indicating differential GC-dependent selection.

Conclusions:

• An entropy-cost trade-off explains how organisms solve the genome encoding problem for a given GC content.
• Empirical boundaries of this trade-off suggest minimum amino acid and codon entropies.
• These findings may impose limits on the GC content observed in natural genomes.