bio-phylogenomicslisted

# Bio Phylogenomics Build marker gene alignments and phylogenetic trees. ## Instructions 1. Validate marker/reference manifests and create a checksum-gated, fixed-seed execution plan: ```bash uv run --no-project python skills/bio-phylogenomics/scripts/run_phylogenomics.py \ markers.tsv --references references.tsv --seed 1729 \ --out results/bio-phylogenomics # Inspect run_manifest.json, then add --execute. ``` The driver restarts only from non-empty stage outputs and normalizes internal support values from either 0–1 or 0–100 notation to `support.tsv` on a 0–1 scale. IQ-TREE `SH-aLRT/UFBoot` labels are emitted as separate `sh_alrt` and `ufboot` rows; mixed scales within one support type fail validation. 2. Extract marker genes or SSU rRNA sequences. 3. Align with MAFFT v7.5+ and trim with trimAl v1.4 (or ClipKIT when phylogenetically-informed trimming is preferred). 4. Build ML trees with support values. Choose by objective first, then leaf count: - Exploratory placement, benchmark iterations, reference-set screening, or any time-bounded analysis: use VeryFastTree v4.0 first, even below ~2,000 taxa. Prefer `VeryFastTree -boot 1000 -threads <n> < alignment.faa > tree.nw` for proteins and add `-nt` for nucleotide alignments. - Final or publication-quality trees up to ~2,000 taxa: IQ-TREE v3 (v3.1.2+) for comprehensive model selection, MAST/GTRpmix, UFBoot/SH-aLRT, and defensible final inference. - Above ~2,000 taxa, or when memory/runtime