New methods for estimating the numbers of synonymous and nonsynonymous substitutions

Area of Science:

• Molecular Evolution
• Bioinformatics
• Computational Biology

Background:

• Accurate estimation of synonymous and nonsynonymous substitutions is crucial for understanding molecular evolution.
• Existing methods, such as Nei and Gojobori (NG), Miyata and Yasunaga (MY), and Li, Wu, and Luo (LWL), have known biases.
• These biases can lead to inaccurate interpretations of evolutionary processes, particularly concerning neutral theory.

Purpose of the Study:

• To develop and evaluate novel, unweighted pathway methods for estimating synonymous and nonsynonymous substitutions per site.
• To compare the accuracy of these new methods against established methods, including NG, MY, LWL, and Pamilo, Bianchi, and Li (PBL).
• To identify limitations and potential biases in current and proposed estimation techniques.

Main Methods:

• Development of new unweighted pathway methods based on Kimura's two-parameter model.
• Extensive computer simulations to assess the accuracy of the new methods and compare them with NG, MY, LWL, and PBL.
• Analysis of estimation biases, including the under/overestimation of substitution and site numbers.

Main Results:

• The NG, MY, and LWL methods consistently overestimate synonymous and underestimate nonsynonymous substitutions due to biased site estimations.
• The PBL method shows improved accuracy over NG, MY, and LWL, but the newly developed methods provide even better estimates.
• New methods offer reasonably accurate estimates of both substitution and site numbers, with a validation mechanism using third-position codon nucleotide substitutions.

Conclusions:

• The developed new methods offer significant improvements in estimating substitution rates compared to previously established techniques.
• The biases identified in older methods (NG, MY, LWL) necessitate re-examination of evolutionary studies relying on them.
• While new methods are generally robust, specific conditions (e.g., mitochondrial genes) may still yield biased estimates, suggesting the utility of alternative approaches like Kondo et al.'s for synonymous substitutions.