LiPP_Benchmark

Introduction

The Lipid–Protein Poses (LiPP) Benchmark is an open-source resource for evaluating protein-lipid docking/co-folding prediction workflows.

Key functions of this repository:

• LiPP dataset: High-quality protein-lipid structure benchmarking datasets.
• Code for reproducing the publication results

LiPP datasets are a subset of the BioDolphin dataset, a comprehensive protein-lipid structural dataset. Check out: https://biodolphin.chemistry.gatech.edu/

Dataset availability:

The LiPP benchmark datasets are located in the data/ directory with the following structure:

CSV Files

• LiPP_csvfiles/: Directory containing IDs(BioDolphin IDs) with their metadata of the LiPP dataset:
  • general_set.csv: LiPP General set for benchmarking
  • test_set.csv: LiPP Test set for benchmarking (A subset of the general set with untrained data)
  • example_set.csv: Small example dataset for testing the workflows

PDB Structures of the LiPP dataset

• pdb_structures_complex/: Contains pdb files of the native protein-lipid complex (named as BioDolphinID.pdb)
• pdb_structures_split/: contains directories named as BioDolphin IDs. Each directory contains the split structures (separating protein and lipid) from the complex structures in pdb_structures_complex/.

Guidance for reproducing the publications

To reproduce the results from our publications, follow the steps below to set up your environment and run the benchmarking workflows.

Prerequisites

1. Before starting, ensure you have the following installed on your system (depending on what you what to test):

• AlphaFold3: https://github.com/google-deepmind/alphafold3
• Chai-1: https://github.com/chaidiscovery/chai-lab
• DiffDock-L: https://github.com/gcorso/DiffDock
• AutoDock Vina: https://vina.scripps.edu/
• RoseTTAFold All-Atom: https://github.com/baker-laboratory/RoseTTAFold-All-Atom
• OpenStructure: https://openstructure.org/install
• ProteinCartography: https://github.com/Arcadia-Science/ProteinCartography

2. Specify the paths for these tools in the config.yaml file in the root directory of this repository

3. Install BioDolphin_vr1.1.csv from https://biodolphin.chemistry.gatech.edu/download and place it under /data

Run the tools:

• scripts/run_tools: Contains scripts for running each tool. See the README.md files in each directory for detailed instructions.

Analyze the results:

• scripts/analyze: Contains scripts for analyzing the tools. See the README.md files in each directory for detailed instructions. The main analyses of the publication can be found in Evalaution/.

Others:

• scripts/filtering: Contains scipts to filter BioDolphin datasets to obtain high quality complexes in LiPP.