lxlsu / drugVQA

Predicting Drug Protein Interaction using Quasi-Visual Question Answering System

Predicting Drug Protein Interaction using Quasi-Visual Question Answering System

Supporting Information for the paper "Predicting Drug Protein Interaction using Quasi-Visual Question Answering System"

DrugVQA is a multimodel learning method combining a dynamic attentive convolutional neural network to learn fixed-size represen-tations from the variable-length distance maps and a self-attentional sequential model to automatically extract semantic features from the linear notations.

Dataset

All data used in this paper are publicly available and can be accessed here: DUD-E, BindingDB-IBM dataset, Human dataset and protein 3D structure.

Demo Instructions

All default arguments for demo are provided in the dataPre.py. Run main.py

Usage

To run the training procedure,

1. Install requirements.txt to set up the envirnoment.
2. Run the main.py to train and test the model.