TorchFold Inference

A structure inference project based on TorchFold.

---

Requirements

• Python 3.11
• Conda

---

Installation

1. Create Environment

conda create -n TorchFold_inference_env python=3.11
conda activate TorchFold_inference_env

2. Install Dependencies

conda install gxx_linux-64 gxx_impl_linux-64 gcc_linux-64 gcc_impl_linux-64=13.2.0

3. Install HMMER

mkdir /path/to/hmmer_build /path/to/hmmer
wget http://eddylab.org/software/hmmer/hmmer-3.4.tar.gz -P /path/to/hmmer_build
cd /path/to/hmmer_build
tar -zxf hmmer-3.4.tar.gz
cd hmmer-3.4
./configure --prefix=/path/to/hmmer
make -j8
make install

4. Install AlphaFold 3.0.1

git clone https://github.com/google-deepmind/alphafold3.git
cd alphafold3
git checkout v3.0.1
pip install -r dev-requirements.txt
pip install --no-deps .
build_data

Download Required Databases:

Download the required AlphaFold databases to your directory :

bash fetch_databases.sh /path/to/database/dir
cd ..

5. Install TorchFold

git clone https://github.com/Mingchenchen/TorchFold.git
cd TorchFold/TorchFold_inference/torchfold
pip install .

---

Running Inference

1. Configure Paths: Edit scripts/env.sh run.sh and modify the path configurations according to your environment.

2. Run Inference:

bash run.sh

---