Home

Awesome

Code for the paper: "PIGNET: A physics-informed deep learning model toward generalized drug-target interaction predictions" by Seokhyun Moon, Wonho Zhung, Soojung Yang, Jaechang Lim, Woo Youn Kim

Table of Contents

Install Dependencies

PIGNet needs conda environment. After installing conda,

you can manually install the necessary packages for executing our code. The package list is shown as follows

or you can just execute the followings in the command line.

./dependencies
conda activate pignet

Running

Data

You can generate the data.

Train

You should download and process the data with data/download_test_data.sh.
Then, Trainig can be done with run.sh.
This process will use the data in ./data/pdbbind_v2019/*.

Model benchmark with CASF2016 benchmark and csar

You should download and process the data with data/download_test_data.sh.
After that, you can use the execution commands for the benchmark are shown in benchmarks/*/test.sh.

Citing this work

@article{moon2022pignet,
  title={PIGNet: A physics-informed deep learning model toward generalized drug-target interaction predictions},
  author={Moon, Seokhyun and Zhung, Wonho and Yang, Soojung and Lim, Jaechang and Kim, Woo Youn},
  journal={Chemical Science},
  year={2022},
  publisher={Royal Society of Chemistry}
}