Multiple Factors Drive Replicating Strand Composition Bias in Bacterial Genomes

Area of Science:

• Genomics
• Bioinformatics
• Microbial Genetics

Background:

• Chargaff's second parity rule (PR2) describes composition bias in sequenced genomes, often linked to microbial genome replication.
• The precise drivers of this strand composition bias remain debated.
• Understanding these drivers is crucial for interpreting genomic data and evolutionary processes.

Purpose of the Study:

• To investigate the multifaceted genomic factors influencing strand composition bias in microbial genomes.
• To determine the relative contributions of various genomic features to this bias.
• To clarify the relationship between replication, transcription, and genomic composition bias.

Main Methods:

• An integrative analysis of genomic features was conducted across a large dataset of 1111 microbial genomes.
• Statistical analyses were employed to assess relationships between composition bias and features like GC content, genome size, gene density, and Clusters of Orthologous Groups (COG) functional categories.
• The study compared bias strength across different lifestyles (obligate intracellular vs. free-living) and microbial phyla.

Main Results:

• Obligate intracellular bacteria exhibited stronger composition bias than free-living species.
• Fusobacteria and Firmicutes phyla showed the highest average bias.
• Negative correlations were observed with GC content, genome size, and rearrangement frequency.
• Positive correlations were found with gene density and specific COG categories (D, F, J, L, V).
• Gene density emerged as the most significant factor, strongly associating transcriptional bias with strand composition bias.

Conclusions:

• Strand composition bias in microbial genomes is a complex phenomenon driven by multiple factors with varying importance.
• Gene density, reflecting transcriptional activity, plays a dominant role.
• The findings provide a more comprehensive understanding of genomic composition bias beyond simple replication-associated models.