pku-embl/metagenomics-skills/bigscape

BiG-SCAPE

Upstream: medema-group/BiG-SCAPE

Citation

Navarro-Muñoz, J. C. et al. A computational framework to explore large-scale biosynthetic diversity. Nat. Chem. Biol. 16, 60–68 (2020). https://doi.org/10.1038/s41589-019-0400-9

Draisma, A. et al. BiG-SCAPE 2.0 and BiG-SLiCE 2.0: scalable, accurate and interactive sequence clustering of metabolic gene clusters. Nat. Commun. (2026). https://doi.org/10.1038/s41467-026-68733-5

Analytical thinking

BiG-SCAPE groups antiSMASH regions into BGC families via similarity networks. Requires consistent antiSMASH outputs. Family membership ≠ identical chemistry — use for prioritization and diversity plots.

How to run

BiG-SCAPE 2 uses subcommands. Supply a curated directory of compatible antiSMASH region GenBanks; do not point it at mixed whole-genome and region GBKs.

set -Eeuo pipefail
input="${1:?usage: cluster-bigscape.sh ANTISMASH_GBK_DIR OUTPUT_DIR}"
out="${2:?usage: cluster-bigscape.sh ANTISMASH_GBK_DIR OUTPUT_DIR}"
threads="${THREADS:-16}"
test -d "$input" || exit 2
find "$input" -type f -name '*.gbk' -print -quit | grep -q . || exit 2
log_dir="${out}.logs"
test ! -e "$out" && test ! -e "$log_dir" || {
  echo "use new output and log paths" >&2; exit 2;
}
mkdir -p "$(dirname "$out")" "$log_dir"
 
bigscape cluster --input-dir "$input" --output-dir "$out" \
  --cores "$threads" --label study --include-singletons \
  >"$log_dir/bigscape.stdout.log" 2>"$log_dir/bigscape.stderr.log"
find "$out" -type f -name '*.db' -print -quit | grep -q .
find "$out" -type f \( -name '*.tsv' -o -name '*.csv' \) -print -quit | grep -q .

MIBiG or custom references must be provided through the version-matched v2 reference options and recorded as a frozen release. BiG-SCAPE 1 flags such as -i/-o/-c are not interchangeable with this interface.

Validation, recovery, and provenance

Require the SQLite database, logs, network/GCF tables, and HTML or configured machine-readable outputs. Reconcile every input region as included, excluded, or failed; inspect antiSMASH version compatibility, duplicate regions, singleton policy, distance/GCF cutoffs, class partitioning, and reference handling. Test cutoff robustness for conclusions based on family counts or novelty.

Use fresh outputs after changing inputs, references, config, or release. Record BiG-SCAPE/Pfam versions, config and cutoff set, antiSMASH and MIBiG releases, input manifest/checksums, command, logs, database/output checksums, and exported network schema. A GCF is a computational similarity cluster, not identical chemistry or activity.

Related skills

antismash · microbial-mining · bakta · visualization