williamgjn/biocreativevii_drugprot

A Sequence Labeling Framework for Extracting Drug-Protein Relations

NLM-NCBI team @ BioCreative VII DrugProt Track

---

This repo contains the source code for extracting drug-protein relations in BioCreative VII DrugProt Track. We propose a novel sequence labeling framework to the drug-protein relation extraction. Our method achieves the top performance in the BioCreative VII DrugProt Track. Please refer to our paper for more details:

• Ling Luo, Po-Ting Lai, Chih-Hsuan Wei, Zhiyong Lu. A sequence labeling framework for extracting drug–protein relations from biomedical literature. Database, 2022, Volume 2022: baac058.

Dependency package

The codes have been tested using Python3.7 on CentOS and uses the following dependencies on a CPU and GPU:

• TensorFlow 2.3.0
• stanza 1.2.2
• transformers 4.8.1

Model preparation

To run this code, you need to first download the model file ( the best single model, i.e., BioM-ELECTRAL with P->D), then unzip and put the model folder into the root folder.

Extracting drug-protein relations from free text

You can use our trained model to extract drug-protein relations from biomedical texts by the /src/DrugProt_Tagging_PD.py file.

The file requires 2 parameters:

• --input, -i, help="input file"
• --output, -o, help="output file to save the extracted relations"

The input file need to provide the text and named entity recognition (NER) information. There is one example in the /example/ folder.

Example:

$ python DrugProt_Tagging_PD.py -i ../example/example_input.tsv -o ../example/example_out.tsv