High-throughput identification of receptor–peptide contact residues from PDB/mmCIF structures (Python 3, Biopython, multithreaded)
extract_binding_site.py is a command-line utility that locates all receptor
residues in close contact with peptide chains for any number of PDB/mmCIF
structures (optionally .gz-compressed).
It assumes the largest protein chain is the receptor and treats all other
protein chains as peptides. Results are written to a single tidy CSV file.
| Key points | |
|---|---|
| Fast & parallel | Uses Python ThreadPoolExecutor; scales to hundreds of structures in minutes |
| Flexible input | Accepts individual files, unix‐style globs, or directories (recurses automatically) |
| Robust parsing | Handles .pdb, .cif, and their .gz versions via Biopython |
| Clean output | Produces a 6-column CSV ready for downstream analysis or visualisation |
| Pure Python 3 | Only external dependencies are biopython, numpy (pulled by Biopython), and tqdm for a neat progress bar |
# 1. Clone the repo
git clone https://github.com/SidSin0809/receptor-peptide-interface-mapper.git
# 2. Install requirements
pip install -r requirements.txt
Quick start
# Example: process all *.pdb files in current & nested folders,
# using 8 threads and a 4 Å cut-off
python extract_binding_site.py "**/*.pdb" \
--cutoff 4.0 \
--threads 8 \
--out peptide_interfaces.csv
CLI options
| Flag | Default | Description |
| ----------- | ------- | -------------------------------------------- |
| `pdb ...` | (none) | File paths, globs, or directories to analyse |
| `--cutoff` | 4.0 | Atom–atom distance threshold (Å) |
| `--threads` | CPU | Number of worker threads (≤ CPU cores) |
| `--out` | csv | Output filename |
Citation If this tool contributes to academic work, please cite:
Singh S. (2025) Receptor–Peptide Interface Mapper. GitHub repository.
https://github.com/SidSin0809/receptor-peptide-interface-mapper
