🧬 PepKit

A Python Toolkit for Peptide Modeling, Analysis, and Benchmarking

PepKit is a modular, peptide-centric Python toolkit designed to support end-to-end computational peptide workflows, from sequence processing and descriptor calculation to structural modeling, confidence assessment, and docking-oriented analysis.

The package is built with reproducibility, scalability, and interoperability in mind, making it suitable for:

• Peptide–protein interaction studies
• Machine-learning–ready dataset construction
• Structural bioinformatics and docking benchmarks
• Large-scale peptide screening and analysis pipelines

✨ Key Features

1️⃣ Sequence I/O & Standardization

• Convert between FASTA and SMILES formats (fasta_to_smiles, smiles_to_fasta).
• Validate and standardize peptide sequences (canonical residues, charge models).
• Batch processing for lists and pandas DataFrames.

2️⃣ Physicochemical Descriptors & Clustering

• Compute peptide-level descriptors (molecular weight, charge, hydrophobicity, pI).
• Generate descriptor tables for ML pipelines.
• Cluster peptide libraries based on sequence or chemical similarity.

3️⃣ Structural Modeling & Confidence Metrics

• Post-process AlphaFold / AlphaFold-Multimer outputs.
• Compute confidence metrics:
  • pLDDT (global, peptide, interface)
  • PAE (interface-aware)
  • pTM / ipTM / composite pTM
  • pDockQ, pDockQ2, MPDockQ
• Optional DockQ evaluation against experimental structures.

4️⃣ Peptide–Protein Dataset Construction

• Automated querying of RCSB PDB for peptide–protein complexes.
• Heuristic peptide-chain detection and interface extraction.
• CSV + FASTA export for benchmarking and ML.

5️⃣ Docking & Benchmark Pipelines

• Prepare inputs for docking and scoring workflows.
• Integrate predicted and experimental metrics in unified tables.
• Designed to interoperate with Rosetta, AlphaFold, and downstream scoring tools.

📦 Installation

Install from PyPI

pip install pepkit

Development installation

git clone https://github.com/Vivi-tran/PepKit.git
cd PepKit
pip install -e .

🚀 Quickstart

This section shows how to get started with PepKit in just a few lines of code, covering the most common peptide-centric tasks.

🔹 Sequence Conversion

Convert between peptide FASTA and SMILES representations:

from pepkit.conversion import fasta_to_smiles, smiles_to_fasta


# FASTA → SMILES
seq = "ACDEFGHIK"
smiles = fasta_to_smiles(seq)
print("SMILES:", smiles)

# SMILES → FASTA
back_seq = smiles_to_fasta(smiles)
print("FASTA:", back_seq)

📚 Documentation

Full documentation is available at:

👉 https://pepkit.readthedocs.io/en/latest/

License

This project is licensed under MIT License - see the License file for details.