OPM Align - Protein Structure Matcher

A bioinformatics tool for finding the closest structural and sequence matches for new protein structures in the OPM (Orientations of Proteins in Membranes) database.

Overview

OPM Align combines sequence alignment (BLAST) with structural superposition (USalign) to identify and classify protein matches. It's particularly useful for transmembrane protein analysis, automatically identifying the subunit with the most transmembrane regions for optimal matching.

Features

• Sequence alignment using NCBI BLAST for initial screening
• Structural superposition using USalign for 3D comparison
• Transmembrane detection via UniProt API integration
• Match classification into 5 types based on identity, overlap, and RMSD
• Cross-platform support (Linux/macOS)
• Multi-threaded processing for faster analysis

Requirements

• Python 3.8+
• NCBI BLAST 2.15.0+
• USalign
• PostgreSQL (for OPM database access)

Installation

1. Clone the repository

git clone git@github.com:oreoqke/opm_align.git
cd opm_align

2. Install NCBI BLAST

Download from: https://blast.ncbi.nlm.nih.gov/doc/blast-help/downloadblastdata.html

macOS (Homebrew):

brew install blast

Linux (Ubuntu/Debian):

sudo apt-get install ncbi-blast+

Verify installation:

blastp -version

3. Install USalign

Download from: https://zhanggroup.org/US-align/

Either add to your PATH or place the USalign executable in the project directory. The tool will automatically search for it.

4. Set up Python environment

# Create virtual environment
python3 -m venv env

# Activate (Linux/macOS)
source env/bin/activate

# Install dependencies
pip install -r requirements.txt

5. Configure database connection

Edit get_all_pdbid.py to specify your OPM database name and credentials if needed.

Usage

Basic usage

python main.py -i new_pdb.txt -o result.csv -d results

Options

Option	Default	Description
-i, --input	new_pdb.txt	File containing list of PDB IDs to match
-o, --output	result.csv	Output file for sequence alignment results
-d, --output-dir	results	Directory for all output files

Input format

Create a text file with one PDB ID per line:

8A9Y
7XYZ
6ABC

Output files

File	Description
results.csv	Complete USalign results
selection.csv	Classified matches with type annotations
top_matches.csv	Best match per PDB (lowest RMSD)
to_align.txt	Pairs selected for structural alignment
bad_matches.txt	PDB IDs with low sequence similarity (<15%)

Match classification

Type	Description
1	Same protein (high identity, overlap, low RMSD)
2	High identity but lower structural overlap
3	Same family (medium identity and overlap)
4	Family match (lower identity, same overlap)
0	No match (filtered out)

Cleanup utility

Use the cross-platform cleanup script to remove generated files:

python cleanup.py fasta   # Remove FASTA files
python cleanup.py log     # Remove log files
python cleanup.py db      # Remove BLAST database files
python cleanup.py all     # Remove all results

Project structure

opm_align/
├── main.py                  # Main entry point
├── align_structures.py      # USalign integration & classification
├── transmembrane_region.py  # UniProt transmembrane detection
├── get_all_fasta.py         # FASTA sequence downloader
├── get_all_pdbid.py         # OPM database connector
├── cleanup.py               # Cross-platform cleanup utility
└── requirements.txt         # Python dependencies

Authors

• Alexey Kovalenko (@oreoqke)
• Stanislav Cherepanov

License

MIT License - see LICENSE for details.

References

• NCBI BLAST documentation
• USalign
• OPM Database