HYphlow: bringing end-to-end flow to HyPhy selection analyses, from taxon reconciliation to trait-based foreground assignment

HYphlow, a Python graphical workflow for HyPhy selection analyses, improves consensus foreground assignment precision to 0.928 versus 0.891 (parsimony) and 0.907 (maximum likelihood) in simulations.

• Why it matters: Manual steps for reconciling taxa, assigning foreground branches, running HyPhy, and summarizing results become harder to standardize as datasets and hypotheses grow.
• What they did: HYphlow integrates taxon reconciliation, ancestral-state reconstruction (parsimony and maximum likelihood), alternative foreground annotations, local or HPC HyPhy execution, and structured result extraction; it used an 80 name-similarity threshold in simulations.
• The result: Consensus foreground assignment increased precision but reduced recall to 0.875 versus 0.921 and 0.918, and HYphlow produced selection results consistent with published cryptochrome (pigeons and doves) and visual opsin (skinks) studies.

AI summary of the abstract · may contain errors · not medical advice · read the paper.