ClipKIT: A multiple sequence alignment trimming software for accurate phylogenomic inference.

Area of Science:

• Computational Biology
• Phylogenetics
• Bioinformatics

Background:

• Highly divergent sites in multiple sequence alignments (MSAs) can negatively impact phylogenetic inference due to homology inference errors and substitution saturation.
• Existing alignment trimming strategies focus on removing uninformative sites, but recent findings suggest this can worsen phylogenetic accuracy.
• There is a need for alternative alignment trimming strategies to improve phylogenetic inference.

Purpose of the Study:

• To introduce ClipKIT, a novel alignment trimming software.
• To develop an alignment trimming strategy that identifies and retains phylogenetically informative sites, specifically parsimony-informative sites.
• To evaluate the efficacy of ClipKIT against existing trimming methods.

Main Methods:

• ClipKIT was developed to identify and retain parsimony-informative sites within MSAs.
• Phylogenetic accuracy and support were assessed using 14 different alignment trimming strategies, including ClipKIT.
• The evaluation utilized nearly 140,000 alignments across diverse evolutionary histories.

Main Results:

• Phylogenies inferred from ClipKIT-trimmed alignments demonstrated high accuracy and robustness.
• ClipKIT significantly improved phylogenetic inference compared to other trimming methods across various datasets.
• The use of ClipKIT resulted in time savings during phylogenetic analysis.

Conclusions:

• ClipKIT offers an effective alignment trimming strategy by focusing on retaining parsimony-informative sites.
• The method of identifying and retaining parsimony-informative sites provides a robust framework for alignment trimming.
• ClipKIT consistently outperforms traditional trimming methods, offering accurate, robust, and time-saving phylogenetic inference.