plantiSMASH database

Code pertaining to the plantiSMASH database setup

LICENSE

https://www.gnu.org/licenses/agpl-3.0

Master plantgeneclusters file

The original plantiSMASH precalculated cluster data are stored per species in directories such as:

/local/data/plantismash/plantismash-data/precalc/v2/<species_id>/plantgeneclusters.txt

where <species_id> is something like Abeliophyllum_distichum_GCA_043235775.1.

Each plantgeneclusters.txt file has the following tab-separated columns:

1. accession – sequence accession (e.g. NC_003070.9)
2. description – sequence description (e.g. Arabidopsis thaliana chromosome 1 sequence)
3. cluster_id – cluster identifier (e.g. c1, c2, c3)
4. cluster_class – predicted BGC class (e.g. polyketide-alkaloid, saccharide, cyclopeptide)
5. genes – semicolon-separated list of gene IDs
6. proteins – semicolon-separated list of protein accessions

To facilitate downstream analyses, we provide a script to build a single master table that concatenates all plantgeneclusters.txt files and adds the species ID as the first column.

Repository layout

• scripts/build_plantgeneclusters_master.sh – script that builds the master table
• data/plantgeneclusters_v2.all.tsv – concatenated table of all plantiSMASH v2 precalculated clusters

Rebuilding the master table

By default, the script assumes that the v2 precalc data are stored at:

/local/data/plantismash/plantismash-data/precalc/v2/

where each species directory contains a plantgeneclusters.txt file.

To rebuild the master table:

cd /path/to/plantismash-database
./scripts/build_plantgeneclusters_master.sh

Test

Count how many BGCs

tail -n +2 data/plantgeneclusters_v2.all.tsv | wc -l

There should be 30426 BGCs in version 2

Create a mapping JSON file

Like the one available on MITE for MIBIG: https://github.com/mite-standard/mite_data/blob/dev/mite_data/mibig/mibig_proteins.json

JSON format

The JSON file (data/plantismash_v2_clusters.json) has the following structure:

{
  "Andrographis_paniculata_GCF_009805555.1/#cluster-1": {
    "accession": "NC_003070.9",
    "description": "Arabidopsis thaliana chromosome 1 sequence",
    "cluster_id": "c1",
    "cluster_class": "polyketide-alkaloid",
    "genes": ["AT1G01960", "AT1G01970", "..."],
    "proteins": ["NP_171698.1", "NP_171699.1", "..."]
  },
  "Andrographis_paniculata_GCF_009805555.1/#cluster-2": {
    ...
  }
}

The JSON keys correspond to the path and anchor part of the plantiSMASH URL:

<species_id>/#cluster-<N>

For example, the key: Andrographis_paniculata_GCF_009805555.1/#cluster-1 corresponds to the web page: https://plantismash.bioinformatics.nl/precalc/v2/Andrographis_paniculata_GCF_009805555.1/#cluster-1

Cluster numbers (cluster-1, cluster-2, etc.) are derived from the numeric part of the cluster_id field (c1, c2, …).

Rebuilding the JSON file

cd /path/to/plantismash-database
./scripts/build_cluster_json.py