Bioinformatics Sequence Retrieval Scripts

A collection of Python utilities for fetching protein and nucleotide sequences from public databases (UniProt and GenBank/NCBI).

Scripts

1. retrieve_records_by_uniprot_ids.py

Fetches protein sequences from UniProt REST API using a list of UniProt accession IDs.

Features:

• Reads UniProt IDs from input file
• Fetches FASTA records from the UniProt REST API
• Writes results to output FASTA file
• Automatic output directory creation
• Progress tracking and summary statistics
• Error handling with detailed logging

Usage:

python scripts/retrieve_records_by_uniprot_ids.py

Input: input/uniprot_ids.txt (one ID per line)
Output: output/results.fasta

Example input file:

P12345
A0A024B7W1
Q9Y5K6

---

2. retrieve_records_by_genebank_id.py

Fetches protein sequences from GenBank (NCBI) using a list of GenBank accession IDs.

Features:

• Reads GenBank IDs from command-line argument
• Fetches FASTA records from NCBI Entrez API
• Writes results to output FASTA file (default: output/results.fasta)
• Automatic output directory creation
• Prevents accidental file overwriting
• Progress tracking and summary statistics
• Error handling with detailed logging

Usage:

python scripts/retrieve_records_by_genebank_id.py <input_file> [output_file]

Examples:

# Use default output location
python scripts/retrieve_records_by_genebank_id.py input/genbank_ids.txt

# Specify custom output location
python scripts/retrieve_records_by_genebank_id.py input/genbank_ids.txt output/custom_results.fasta

Input: Comma or space-separated GenBank IDs (one per line or multiple per line)
Output: output/results.fasta (or custom path)

Example input file:

NM_000001
NC_000001
AF123456

---

Project Structure

bioinformatics/
├── scripts/
│   ├── retrieve_records_by_uniprot_ids.py
│   └── retrieve_records_by_genebank_id.py
├── input/
│   ├── uniprot_ids.txt
│   └── genbank_ids.txt
├── output/
│   └── results.fasta
└── README.md

---

Requirements

• Python 3.6+
• Dependencies:
  • requests - For HTTP requests
  • biopython - For sequence parsing and writing
  • Bio.Entrez - For NCBI database access (included in biopython)

Install dependencies:

pip install requests biopython

---

Configuration

UniProt Script

The UniProt script uses hardcoded paths:

• Input: input/uniprot_ids.txt
• Output: output/results.fasta

GenBank Script

The GenBank script requires command-line arguments:

python scripts/retrieve_records_by_genebank_id.py <input_file> [output_file]

The NCBI Entrez email is set to: erbayyigit@gmail.com
(Modify in the script if needed)

---

Features

✅ Automatic directory creation
✅ Error handling and validation
✅ Input file existence checking
✅ Duplicate file prevention (GenBank script)
✅ Progress tracking during retrieval
✅ Summary statistics after completion
✅ Detailed logging to stderr for errors

---

Error Handling

Both scripts include:

• Input validation: Checks if input file exists
• Output validation: Prevents overwriting existing files (GenBank)
• API error handling: Catches HTTP errors and parsing exceptions
• Polite rate limiting: 0.1 second delay between API requests

---

Output Format

All output is in FASTA format:

>sequence_id description
MKVLWAALLV...

Each sequence retrieved is appended to the output file.

---

Notes

• The UniProt script uses fixed file paths. Modify the script if you need different paths.
• The GenBank script accepts input filename as a command-line argument.
• Both scripts create the output/ directory automatically if it doesn't exist.
• Rate limiting is set to 0.1 seconds between requests to be respectful to public APIs.
• An NCBI Entrez email is required for GenBank queries. Update it in the script if needed.

---

Legacy Scripts

Older edirect utilities are available in the edirect-utils/ directory:

• efetch-proteins-by-id.py - Original NCBI protein fetcher
• efetch-proteins-by-id-and-analyze.py - NCBI fetcher with protein analysis features

These have been superseded by the modern scripts in the scripts/ directory with improved error handling and user experience.