Modeling compositional dynamics based on GC and purine contents of protein-coding sequences

Area of Science:

• Genomics
• Molecular Evolution
• Bioinformatics

Background:

• Understanding genome compositional dynamics is crucial for molecular evolution studies.
• Quantifying theoretical compositional variations across diverse genomes remains challenging.

Purpose of the Study:

• To propose two models for predicting the compositional dynamics of protein-coding sequences.
• To account for mutation and selection effects at different codon positions.
• To utilize GC and purine content as key compositional parameters.

Main Methods:

• Developed two theoretical models for nucleotide, codon, and amino acid composition.
• Models do not require homologous sequences or alignments.
• Evaluated models on a large dataset of Archaea, Bacteria, and Eukarya protein-coding sequences.

Main Results:

• Achieved consistent theoretical compositions across all evaluated sequences.
• Demonstrated that GC and purine content largely determine nucleotide, codon, and amino acid compositions.
• Validated the models' ability to predict compositional dynamics.

Conclusions:

• Nucleotide, codon, and amino acid compositions are primarily governed by GC and purine content.
• Deviations from expected compositions may indicate signatures of mutation and selection.
• Suggests a link between DNA replication/repair mechanisms and compositional signatures.