Chemical Probes in Scientific Literature: A Pipeline for Target-Disease Evidence

This repository contains the main datasets generated for the manuscript, "Chemical Probes in Scientific Literature: Expanding and Validating Target-Disease Evidence." The datasets were derived using a Natural Language Processing (NLP) pipeline to systematically analyze open-access scientific literature for evidence of target-disease (T-D) associations mediated by chemical probes. This repository provides the primary results and aggregated datasets underpinning the manuscript's findings. This includes the dataset used for the pilot study, the main systematic dataset containing Probe-Target-Disease (P-T-D) triples annotated with Open Targets data, the novel target-disease pairs identified via the study's scoring scheme, and aggregated files for specific entities and filtering steps. These data can be used to reproduce the analyses in the manuscript, explore the landscape of chemical probe-based evidence, and identify novel T-D associations for further investigation. The complete workflow, including all intermediate files tracking the extraction process, can be reproduced by executing the main pipeline available on GitHub at: https://github.com/chembl/chemical_probes_lit File Descriptions This record contains the datasets extracted from the systematic pipeline and the pilot study evaluating chemical probes, targets, and diseases: pilot_triplets.tsv: Articles found in the pilot study with an SGC chemical probe, target, and disease. The file reports all triples found in articles. probes_HQ.tsv: High-quality probes dictionary used in the automated approach. ner_all_triplets.tsv: First step for the systematic approach with all articles annotated with 3 entities [any chemical, target, disease] if they are found within the same sentence. ner_diseases.tsv: All diseases found in the main dataset with the corresponding ontology disease id (diseasetId), the disease’s name (diseaseName), the total number of articles (pmids) they appear in, the number of unique targets they are linked to (targets), and the number of unique chemical probes (probes). They are sorted by frequency with respect to the pmids. ner_probes.tsv: All chemical probes found in the main dataset with the corresponding ChEMBL id (probeId), the chemical probe name (probeName), the total number of articles (pmids) they appear in, the number of unique diseases they are linked to (diseases), and the number of unique targets (targets). They are sorted by frequency with respect to the pmids. ner_targets.tsv: All targets found in the main dataset with the corresponding Ensembl target id (targetId), the target’s gene symbol (geneName), the total number of articles (pmids) they appear in, the number of unique diseases they are linked to (diseases), and the number of unique chemical probes (probes). They are sorted by frequency with respect to the pmids. ner_target_disease-pairs.tsv: all unique target-disease (T-D) pairs from the main dataset sorted by frequency with respect to the number of articles they appear in articles. ner_probes_triplets_ptpairs.tsv: Main dataset for the systematic approach with all articles annotated with 3 entities [chemical probe, target, disease] if they are found within the same sentence. The found target is one of the approved targets for the specific chemical probe. ner_probes_triplets_ptpairs_evd.tsv: Main dataset for the systematic approach with 3 entities and existing T-D evidence from the OpenTarget Platform (OTP) if available. It also includes details on dated evidence if found, OTP therapeutic area for disease entity, and OTP known drugs/clinical phase for T-D pairs. ner_probes_early_literature.tsv: The subset contains the T-D pairs for which chemical probes literature evidence was found before any other non-literature evidence from Open Targets. ner_probes_triplets_novel.tsv: The subset contains the novel T-D pairs following the strict filter criteria: selected pairs must be supported by recent evidence (first probe publication <5 years ago), not overly studied (≤ 20 articles) and must have been tested with multiple tools (≥2 distinct chemical probes), yet lack any approved drugs or active clinical programs. disease_term_blacklist.tsv: A compilation of disease identifiers excluded from the analysis. These terms were flagged during manual curation as systematic false positives within the NER dataset, primarily due to ambiguous abbreviations that led to incorrect disease assignments.