TCS: a new multiple sequence alignment reliability measure to estimate alignment accuracy and improve phylogenetic...

Area of Science:

• Bioinformatics
• Computational Biology
• Evolutionary Biology

Background:

• Multiple sequence alignment (MSA) is crucial for biological sequence analysis, particularly in homology and evolutionary modeling.
• The accuracy of MSAs directly impacts the reliability of downstream analyses like phylogenetic tree reconstruction.
• Existing methods for assessing MSA quality have limitations.

Purpose of the Study:

• To introduce and evaluate the transitive consistency score (TCS) as a method to identify reliable portions of MSAs.
• To demonstrate the utility of TCS in improving phylogenetic tree reconstruction.
• To present a novel, lossless approach to MSA refinement by overweighting trustworthy columns.

Main Methods:

• Developed TCS, an extension of the T-Coffee scoring scheme, for local MSA evaluation.
• Utilized BAliBASE and PREFAB reference alignments to validate TCS's accuracy in identifying reliable MSA regions.
• Implemented a novel column weighting strategy within the T-Coffee framework, using pairwise alignment libraries.
• Compared TCS performance against Heads-or-Tails, GUIDANCE, Gblocks, and trimAl.

Main Results:

• TCS effectively identifies the most reliable regions within MSAs.
• Phylogenetic tree reconstruction accuracy was significantly improved using TCS on both simulated and empirical yeast datasets.
• The novel column weighting method proved a lossless alternative to traditional site filtering.
• TCS outperformed existing methods in estimating structural accuracy and reconstructing phylogenetic trees.

Conclusions:

• TCS offers a robust method for assessing MSA quality and enhancing downstream analyses.
• The transitive consistency score provides a valuable tool for improving the accuracy of phylogenetic inference.
• The T-Coffee framework, extended with TCS, facilitates more reliable biological sequence modeling.