SRH

SRH assesses violations of stationarity, reversibility, and homogeneity in substitution models of sequence evolution and quantifies their impact on phylogenetic inference.

Key Features:

• Maximal Matched-Pairs Tests: SRH employs maximal matched-pairs tests to evaluate homogeneity across sequence partitions.
• Impact Assessment: The software quantifies the impact of SRH violations on phylogenetic inference, focusing on effects on tree topologies and edge lengths and comparing trees inferred from all partitions versus SRH-conforming partitions.
• Data Analysis Capability: SRH analyzes a dataset comprising 3,572 partitions from 35 published phylogenetic datasets to assess the prevalence of model violations.
• Integration with IQ-TREE: SRH provides a testing option within IQ-TREE to identify and allow exclusion of partitions that violate SRH assumptions prior to tree reconstruction.

Scientific Applications:

• Model Validation: Identification of partitions that violate SRH assumptions to inform selection or refinement of substitution models for more accurate phylogenetic trees.
• Improved Phylogenomic Inference: Informing the development and evaluation of substitution models that relax SRH assumptions to enhance phylogenomic analyses.

Methodology:

SRH applies maximal matched-pairs tests, calculates summary statistics, and prepares input files for IQ-TREE analysis; it requires Python 3.6.x or higher with dependencies listed in the srh.py script header and references hardcoded input/output paths under /data/srh/processed_data/SRH_tables/ and /data/srh/processed_data/IQtree/.