#!/usr/bin/env bash
# Fetch the public demo slice: a real GIAB genome and real ClinVar labels, chr22 only.
# Provenance, licences and citations: docs/data.md
set -euo pipefail

OUT_DIR=${OUT_DIR:-data}
CLINVAR_URL=${CLINVAR_URL:-https://ftp.ncbi.nlm.nih.gov/pub/clinvar/vcf_GRCh38/clinvar.vcf.gz}
GIAB_URL=${GIAB_URL:-https://ftp-trace.ncbi.nlm.nih.gov/ReferenceSamples/giab/release/AshkenazimTrio/HG002_NA24385_son/NISTv4.2.1/GRCh38/HG002_GRCh38_1_22_v4.2.1_benchmark.vcf.gz}

for tool in bcftools tabix; do
  command -v "$tool" >/dev/null || {
    echo "$tool is required (brew install bcftools, or apt install bcftools tabix)" >&2
    exit 1
  }
done

mkdir -p "$OUT_DIR"

# Both sources are indexed, so bcftools streams one chromosome instead of downloading a whole
# genome. ClinVar names contigs "22"; GIAB names them "chr22".
echo "==> GIAB HG002 (NA24385) v4.2.1 benchmark, chr22 -> $OUT_DIR/example.vcf.gz"
bcftools view -r chr22 "$GIAB_URL" -Oz -o "$OUT_DIR/example.vcf.gz"
tabix -f -p vcf "$OUT_DIR/example.vcf.gz"

echo "==> ClinVar GRCh38, chr22 -> $OUT_DIR/clinvar.chr22.vcf.gz"
bcftools view -r 22 "$CLINVAR_URL" -Oz -o "$OUT_DIR/clinvar.chr22.vcf.gz"
tabix -f -p vcf "$OUT_DIR/clinvar.chr22.vcf.gz"

echo
echo "Fetched:"
ls -lh "$OUT_DIR/example.vcf.gz" "$OUT_DIR/clinvar.chr22.vcf.gz"
cat <<'NEXT'

Next:
  make pipeline VCF=data/example.vcf.gz          # annotate the GIAB sample (needs a VEP cache)
  make annotate JOB=<job id> VCF=data/example.vcf.gz
NEXT