A novel framework for evaluating the performance of codon usage bias metrics.

Area of Science:

• Genomics
• Molecular Biology
• Bioinformatics

Background:

• Unequal synonymous codon utilization influences key cellular processes like translation and mRNA stability.
• Accurate measurement of codon usage bias (CUB) is critical for understanding gene and genome-level biological impacts.
• Existing CUB metrics lack systematic validation, potentially leading to unreliable research findings.

Purpose of the Study:

• To address the lack of standardization in CUB metric evaluation.
• To identify the most accurate and robust CUB metrics.
• To provide a benchmark for developing and validating new CUB metrics.

Main Methods:

• Development of a novel method to generate synthetic protein-coding DNA sequences based on defined codon usage models.
• Systematic testing of various CUB metrics using these synthetic sequences.
• Evaluation of metric performance based on sequence length, GC content, and amino acid heterogeneity.

Main Results:

• The choice of CUB metric significantly affects statistical significance and measured effect sizes in empirical data.
• Identified specific CUB metrics demonstrating high accuracy and robustness across different sequence characteristics.
• Demonstrated the utility of the benchmark for assessing the performance of existing and novel CUB metrics.

Conclusions:

• Standardized evaluation of CUB metrics is necessary to ensure the reliability of biological discoveries.
• The developed benchmark facilitates the selection of appropriate CUB metrics and aids in the development of improved methodologies.
• This work enhances the rigor and reproducibility of research investigating codon usage bias.