radiyaman/funfold5_template repository overview

⁠FunFOLD5: Template-Based Prediction for Protein–Ligand Interactions

This repository provides a Dockerized version of ** the FunFOLD5-template pipeline**, a template-based protein–ligand binding site prediction tool developed by the McGuffin Group.

The prediction process includes two main stages:

1. Global Binding Site Prediction: FunFOLD5 is used to identify biologically relevant binding sites and ligands based on the template list for the entire protein structure.
2. Local Binding Site Prediction: The protocol is reapplied to the most abundant binding sites to uncover additional ligand-binding templates with enhanced specificity.

⁠🔍 Why Use FunFOLD5?

Compared to previous versions (e.g., FunFOLD3), FunFOLD5 offers several advantages:

• 🚫 No need for model generation – avoids GPU-heavy steps.
• 💻 Lightweight and CPU-friendly – perfect for local machines.
• 🎯 Improved accuracy for predicting local binding sites and ligands.
• 🔍 Identifies templates not only across the whole protein but also for specific binding local sites.

⁠📚 Citation

If you use this tool in your work, please cite the following publication:

⁠⚙️ Requirements

• Docker (latest version recommended)
• Linux system (developed on Ubuntu 20.04)
• PDB-format input MODEL FILE
• BioLip2 database (must be downloaded and mounted locally)

⁠📥 Downloading BioLip2

You can download the BioLip2 database from:

[https://zhanggroup.org/BioLiP/weekly.html⁠)

Use the following script to download the full dataset:

wget https://zhanggroup.org/BioLiP/download/download_all_sets.txt
perl download_all_sets.txt

⁠🧪 How to Run

Prepare your inputs:

• A protein model file in PDB format
• BioLip2 database available, and its path supplied

⁠🔁 Example 1

docker run --rm \
  -e WORKING_FOLDER=/data/work \
  -e MODEL_FILE=T1214.pdb \
  -e BIOLIP_Directory=/data/biolip \
  -v /path/to/working_folder:/data/work \
  -v /path/to/BioLip2:/data/biolip \
  radiyaman/funfold5_template:latest \
  bash /FunFOLD5_files/run_FunFOLD5.sh

⁠🔁 Example 2 (Full Path)

docker run --rm \
  -e WORKING_FOLDER=/data/work \
  -e MODEL_FILE=T1214.pdb \
  -e BIOLIP_Directory=/data/biolip \
  -v /home/recep/Desktop/Docker_FunFOLD/test:/data/work \
  -v /home/recep/Desktop/BioLiP2/BioLiP_updated_set:/data/biolip \
  radiyaman/funfold5_template:latest \
  bash /FunFOLD5_files/run_FunFOLD5.sh

⁠📂 Output Description

All outputs are saved in the FunFOLD5 subfolder within your working directory.

⁠Key Output Files:

• *_ligands.pdb: Ligands used for binding site predictions
• Foldseek_*.dat: Final template lists for each binding site
• *.m8: Foldseek-generated template lists for each site
• all_ligand_list.txt: Complete list of ligands across all templates
• all_ligand_summary_list.txt: Summarized ligand names
• bindingsite*.txt: Binding site Residues for each predicted binding site
• Allbindingsite*.txt: Binding site residues and their neighbours within 5 Å
• *_FN.txt: Predicted residues, ligands, GO annotations, and cluster details and all related information summarised in this gile
• *_lig.pdb: Protein structure with all ligands from the template list
• lig2.pdb: Protein structure with all biologically relevant ligands
• lig3.pdb: Protein structure with all centroid ligands
• summary_3D_sequence_*.txt and summary_TM_sequence_*.txt: E-values and TM-scores for each template

These files together offer structural, functional, and statistical insight into ligand-binding predictions.

⁠👤 Contact

Dr. Recep Adiyaman
School of Biological Sciences
University of Reading

Email: [email protected]⁠

⁠License

MIT License

Copyright (c) 2025 Liam McGuffin, Recep Adiyaman, and The University of Reading

Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

Third-Party Software Notices for FunFOLD5 Docker image⁠⁠