ForeSeqs

ForeSeqs predicts missing sequences and corrects branch lengths in phylogenomic datasets to improve the accuracy of phylogenetic inference.

Key Features:

• Prediction of Missing Sequences: Predicts missing sequence data for organisms by leveraging genes or partitions that contain complete data (per-gene or per-partition) to fill gaps in phylogenomic datasets.
• Correction of Long Branch Lengths: Implements algorithms to correct excessively long branch lengths induced by missing data in phylogenetic trees.
• Evaluation with Empirical and Simulated Alignments: Evaluates performance using systematic removal of sequence data from three empirical and 100 simulated alignments.
• Comparison with Maximum Likelihood Trees: Compares Maximum Likelihood trees inferred from gappy alignments (with predicted sequences) to trees from complete datasets and reports branch length accuracy improvements by one to two orders of magnitude.
• Effect on Robinson-Foulds Distances: Assesses Robinson-Foulds (RF) distances and reports that prediction of missing sequences does not affect RF distances between trees.

Scientific Applications:

• Evolutionary biology: Improves phylogenetic inference and branch length estimation for studies in evolutionary biology.
• Large-scale phylogenomics: Enhances completeness of large-scale genomic datasets with per-gene or per-partition gaps for downstream phylogenetic analyses.
• Species divergence and lineage tracing: Supports analyses sensitive to branch length accuracy, such as species divergence estimation and genetic lineage tracing.

Methodology:

Predicts missing sequences by leveraging complete genes/partitions; applies algorithms to correct long branch lengths; evaluates performance via systematic removal of sequence data from three empirical and 100 simulated alignments and by comparing Maximum Likelihood trees and Robinson-Foulds distances between gappy (with predicted sequences) and complete datasets.