Gaussian-Distributed Codon Frequencies of Genomes

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• DNA uses 64 codons to specify 20 amino acids and a stop signal, exhibiting redundancy.
• Codon usage bias, reflecting genome-specific codon preferences, is a key characteristic.
• Previous models described codon usage with empirical expressions.

Purpose of the Study:

• To propose a statistical model for co-existing codon usage patterns within a genome.
• To investigate if this model explains diverse codon usage across taxa.
• To identify parameters that characterize differences in codon usage frequency/rank relations.

Main Methods:

• Developed a statistical model positing two codon usage compartments: weak (Gaussian) and strong bias.
• Analyzed codon usage frequency/rank relations across a large dataset of genomes.
• Quantified differences using a single parameter: the proportion of the two compartments.

Main Results:

• The model successfully accounts for observed codon usage variations across different taxa.
• A single parameter, the proportion of codon compartments, explains inter-genomic differences.
• Prokaryotic genomes show varied proportions of both compartments; eukaryotic genomes are dominated by the Gaussian compartment.

Conclusions:

• Codon usage bias can be modeled by two co-existing compartments with distinct bias strengths.
• The proportion of these compartments is a key factor in genome evolution and diversity.
• Gaussian codon frequencies suggest multiple evolutionary pressures, while strong bias indicates fewer dominant pressures.