DrugProt Large-Scale Text Mining corpus: Biocreative VII Track 1 - Text mining drug and chemical-protein interactions

This Zenodo contains the BioCreative VII Large scale DrugProt Additional Subtrack abstracts and entity annotations.

Abstracts

• large_scale_abstracts.tsv This file contains plain-­text, UTF8-­encoded, NFC normalized DrugProt PubMed records in a tab­ ‐ separated format. In total 2366081 records are provided, where each line in the fails contains a single PMID, title and abstract separated by tabulators. Due to PubMed inconsistencies, there is a minor percentage of duplicated records. Indeed, we have identified 222 records with different PMID but the same abstract title and body.

Entity mention annotations

• large_scale_entities.tsv. This file contains the automatically labeled mention annotations of chemical compounds and genes/proteins (so-­called gene and protein-related objects as defined during BioCreative V) generated for the Large Scale records. There are 53993602 entity annotations.

Related resources:

•  
• Web
• DrugProt corpus
• Evaluation library
• Online evaluation (CodaLab)
• Relation annotation guidelines
• Gene and protein annotation guidelines
• Chemicals and drugs annotation guidelines
• DrugProt Silver Standard Knowledge Graph
• FAQ
• DrugProt Large Scale Additional SubTrack
• DrugProt Large Scale document collection protocol
• DrugProt Complete PubMed Knowledge Graph