CAFE-MPP: Self-Supervised Learning with Chemistry-aware Fragmentation for Effective Molecular Property Prediction
Ailin Xie, Ziqiao Zhang, Jihong Guan, Shuigeng Zhou
Fudan University, Tongji University
This is the official implementation of CAFE-MPP: " Self-Supervised Learning with Chemistry-aware Fragmentation for Effective Molecular Property Prediction ".
# conda environment
conda create --name CAFE-MPP python=3.8
conda activate CAFE-MPP
# install requirements
conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
conda install pyg -c pyg
conda install -c conda-forge rdkit
conda install -c anaconda cython
# clone the source code
git clone https://github.com/shiokoo/CAFE-MPP.git
cd CAFE-MPP
We provide the preprocessed pre-training fragment dataset used in this project. Besides, You can download the benchmarks MoleculeNet used in this project by running the following command:
cd ./Data
bash download_data.sh
To pre-train the CAFE-MPP, where the configurations and hyperparameters are defined in ./Config/config_pretrain.yaml.
cd ./Pretrain
python trainer.py
To fine-tune the CAFE-MPP, where the configurations and details are can be found in ./Config/config_prediction.yaml.
cd ./Prediction
python setup.py build_ext --inplace # generate .so
python trainer.py
If you find our work is helpful in your research, please cite:
@article{xie2023cafe-mpp,
author = {Xie, Ailin and Zhang, Ziqiao and Guan, Jihong and Zhou, Shuigeng},
title = "{Self-supervised learning with chemistry-aware fragmentation for effective molecular property prediction}",
journal = {Briefings in Bioinformatics},
pages = {bbad296},
year = {2023},
month = {08},
issn = {1477-4054},
doi = {10.1093/bib/bbad296},
url = {https://doi.org/10.1093/bib/bbad296},
eprint = {https://academic.oup.com/bib/advance-article-pdf/doi/10.1093/bib/bbad296/51175671/bbad296.pdf},
}
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent