Latent Target Diffusion

This repository is the implementation of Latent TargetDiff, for my CPSC 483 final project. The final paper can be found here.

This repository is based off of the following two repositories:

https://github.com/guanjq/targetdiff

https://github.com/MinkaiXu/GeoLDM

Script for Producing Main Results:

summary.ipynb

To run the notebook, download all baselines from the folder linked below (in "sampling_results").

Links for GeoLDM autoencoder and baseline samples:

Files to download are located here.

Environment

Dependency

The code has been tested in the following environment:

Create Mamba environment

mamba env create -f environment.yaml
conda activate latenttargetdiff  # note: one still needs to use `conda` to (de)activate environments

---

Target-Aware Molecule Generation

Data

The data used for training / evaluating the model are organized in the data Google Drive folder.

To train the model from scratch, you need to download the preprocessed lmdb file and split file:

• crossdocked_v1.1_rmsd1.0_pocket10_processed_final.lmdb
• crossdocked_pocket10_pose_split.pt

To evaluate the model on the test set, you need to download and unzip the test_set.zip. It includes the original PDB files that will be used in Vina Docking.

Training

Training from scratch

python scripts/train_diffusion.py configs/training.yml

Sampling

Sampling for pockets in the testset

python scripts/sample_diffusion.py configs/sampling.yml --data_id {i} # Replace {i} with the index of the data. i should be between 0 and 99 for the testset.

Evaluation

Evaluation from sampling results

python scripts/evaluate_diffusion.py {OUTPUT_DIR} --docking_mode vina_score --protein_root data/test_set

The docking mode can be chosen from {qvina, vina_score, vina_dock, none}

Note: It will take some time to prepare pqdqt and pqr files when you run the evaluation code with vina_score/vina_dock docking mode for the first time.