Add fetch_refs.sh to download BFVD + ICTV backing data

Idempotent wget-based shell script that pulls the three BFVD
metadata/taxid/lineage TSVs from steineggerlab and the current
ICTV VMR MSL41 xlsx into refs/downloads/. README updated with a
new step 0 that runs before build_bfvd_refs / build_host_table.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

Author: jonn-smith

analysis/hvp_viral_viz/README.md

@@ -70,6 +70,9 @@ External fill (host species, see [`HOST_FILL.md`](HOST_FILL.md)):
 ## Pipeline
 
 ```bash
+# 0. Fetch backing data sources (BFVD TSVs + ICTV VMR xlsx) — one-time
+bash fetch_refs.sh --out-dir refs/downloads
+
 # 1. Build reference joins (one-time)
 PYTHONPATH=.. python -m hvp_viral_viz.build_bfvd_refs
 PYTHONPATH=.. python -m hvp_viral_viz.uniprot_host --source bfvd

analysis/hvp_viral_viz/fetch_refs.sh

@@ -0,0 +1,69 @@
+#!/usr/bin/env bash
+# Fetch the backing data sources required by build_bfvd_refs.py + build_host_table.py.
+#
+# Downloads (idempotent — skips files already present and non-empty):
+#   1. BFVD metadata + taxid + lineage TSVs from steineggerlab.
+#   2. ICTV VMR MSL41 xlsx from ictv.global.
+#
+# Usage:
+#   bash fetch_refs.sh [--out-dir DIR]
+#
+# Then point the build scripts at the downloaded files:
+#   python -m hvp_viral_viz.build_bfvd_refs   --bfvd-dir <out>/bfvd  --out-dir refs
+#   python -m hvp_viral_viz.build_host_table  --refs-dir refs       --ictv-vmr <out>/ictv_vmr_msl41.xlsx
+
+set -euo pipefail
+
+OUT_DIR="./refs/downloads"
+while [[ $# -gt 0 ]]; do
+    case "$1" in
+        --out-dir) OUT_DIR="$2"; shift 2 ;;
+        -h|--help) sed -n '2,15p' "$0"; exit 0 ;;
+        *) echo "unknown arg: $1" >&2; exit 2 ;;
+    esac
+done
+
+mkdir -p "$OUT_DIR/bfvd"
+
+# ---- BFVD ----------------------------------------------------------------
+BFVD_BASE="https://bfvd.steineggerlab.workers.dev/latest"
+BFVD_FILES=(
+    "bfvd_metadata.tsv"
+    "bfvd_taxid.tsv"
+    "bfvd_taxid_rank_scientificname_lineage.tsv"
+)
+
+for f in "${BFVD_FILES[@]}"; do
+    dest="$OUT_DIR/bfvd/$f"
+    if [[ -s "$dest" ]]; then
+        echo "[fetch_refs] skip BFVD $f (already present)"
+        continue
+    fi
+    echo "[fetch_refs] downloading BFVD $f"
+    wget -c -O "$dest.partial" "$BFVD_BASE/$f"
+    mv "$dest.partial" "$dest"
+done
+
+# ---- ICTV VMR ------------------------------------------------------------
+# Canonical filename build_host_table.py looks for: ictv_vmr_msl41.xlsx.
+# Upstream filename embeds release date and may change between MSLs; we
+# rename to the stable local name.
+ICTV_URL="https://ictv.global/sites/default/files/VMR/VMR_MSL41.v1.20260320.xlsx"
+ICTV_DEST="$OUT_DIR/ictv_vmr_msl41.xlsx"
+
+if [[ -s "$ICTV_DEST" ]]; then
+    echo "[fetch_refs] skip ICTV VMR (already present)"
+else
+    echo "[fetch_refs] downloading ICTV VMR MSL41"
+    wget -c -O "$ICTV_DEST.partial" "$ICTV_URL"
+    mv "$ICTV_DEST.partial" "$ICTV_DEST"
+fi
+
+# ---- Summary -------------------------------------------------------------
+echo
+echo "[fetch_refs] done. files in $OUT_DIR/:"
+ls -lh "$OUT_DIR/bfvd/" "$OUT_DIR/"*.xlsx 2>/dev/null || true
+echo
+echo "Next steps:"
+echo "  python -m hvp_viral_viz.build_bfvd_refs  --bfvd-dir $OUT_DIR/bfvd  --out-dir refs"
+echo "  python -m hvp_viral_viz.build_host_table --refs-dir refs --ictv-vmr $ICTV_DEST"