PanGBank-annotator

Introduction

PanGBank-annotator is a workflow for building PANFAM protein clusters from PanGBank pangenome family representatives and annotating the resulting representative sequences.

The current implementation includes the following steps:

1. Resolve a PanGBank collection release and collection (GTDB_all or GTDB_refseq).
2. Fetch all protein families of each pangenome and build a pangenome-family map.
3. Run DIAMOND deepclust, followed by recluster and reassign corrections.
4. Package clusters into PANFAM parquet files and representative FASTA files.
5. Annotate PANFAM80 representatives with InterProScan6, DeepKOALA, eggNOGMapper, and AMRFinder+.
6. Package annotation results and generate a comprehensive multiQC report.

Usage

Note

If you are new to Nextflow and nf-core, please refer to this page on how to set-up Nextflow. Make sure to test your setup with -profile test before running the workflow on actual data.

Detailed run examples are available in docs/usage.md, including stub runs, built-in tests, clustering-only, annotation-only, full runs, custom FASTA input, annotation tool selection, and raw output controls.

Stage-specific details are in docs/clustering.md and docs/annotation.md. Annotation database setup is covered in docs/databases.md.

Technical Notes

• DIAMOND is currently pinned to 2.1.13 for all clustering steps. v2.1.24 fails during reassign, and v2.2.4 temporarily removed reassign. The DIAMOND modules declare both Conda and container execution for this pinned version.

• DeepKOALA is run from a dedicated container image and external model resources. Build the default image from modules/local/deepkoala/Dockerfile and provide the resources directory with --deepkoala_resources.

• Imported InterProScan 6 is tracked as an upstream Git submodule. Until the upstream sequences.db reported issue is fixed, initialize the submodule and apply the local temporary patch before running imported InterProScan:

git submodule update --init --recursive
bin/apply_interproscan6_patches.sh

Citations

An extensive list of references for the tools used by the pipeline can be found in the CITATIONS.md file.