1
|
Chambers BA, Basili D, Word L, Baker N, Middleton A, Judson RS, Shah I. Searching for LINCS to Stress: Using Text Mining to Automate Reference Chemical Curation. Chem Res Toxicol 2024; 37:878-893. [PMID: 38736322 DOI: 10.1021/acs.chemrestox.3c00335] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 05/14/2024]
Abstract
Adaptive stress response pathways (SRPs) restore cellular homeostasis following perturbation but may activate terminal outcomes like apoptosis, autophagy, or cellular senescence if disruption exceeds critical thresholds. Because SRPs hold the key to vital cellular tipping points, they are targeted for therapeutic interventions and assessed as biomarkers of toxicity. Hence, we are developing a public database of chemicals that perturb SRPs to enable new data-driven tools to improve public health. Here, we report on the automated text-mining pipeline we used to build and curate the first version of this database. We started with 100 reference SRP chemicals gathered from published biomarker studies to bootstrap the database. Second, we used information retrieval to find co-occurrences of reference chemicals with SRP terms in PubMed abstracts and determined pairwise mutual information thresholds to filter biologically relevant relationships. Third, we applied these thresholds to find 1206 putative SRP perturbagens within thousands of substances in the Library of Integrated Network-Based Cellular Signatures (LINCS). To assign SRP activity to LINCS chemicals, domain experts had to manually review at least three publications for each of 1206 chemicals out of 181,805 total abstracts. To accomplish this efficiently, we implemented a machine learning approach to predict SRP classifications from texts to prioritize abstracts. In 5-fold cross-validation testing with a corpus derived from the 100 reference chemicals, artificial neural networks performed the best (F1-macro = 0.678) and prioritized 2479/181,805 abstracts for expert review, which resulted in 457 chemicals annotated with SRP activities. An independent analysis of enriched mechanisms of action and chemical use class supported the text-mined chemical associations (p < 0.05): heat shock inducers were linked with HSP90 and DNA damage inducers to topoisomerase inhibition. This database will enable novel applications of LINCS data to evaluate SRP activities and to further develop tools for biomedical information extraction from the literature.
Collapse
Affiliation(s)
- Bryant A Chambers
- Center for Computational Toxicology and Exposure, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, United States
| | - Danilo Basili
- Unilever, Safety and Environmental Assurance Centre (SEAC), Colworth Science Park, Sharnbrook, Bedfordshire MK44 1LQ, U.K
| | - Laura Word
- Center for Computational Toxicology and Exposure, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, United States
| | - Nancy Baker
- Leidos, Research Triangle Park, North Carolina 27711, United States
| | - Alistair Middleton
- Unilever, Safety and Environmental Assurance Centre (SEAC), Colworth Science Park, Sharnbrook, Bedfordshire MK44 1LQ, U.K
| | - Richard S Judson
- Center for Computational Toxicology and Exposure, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, United States
| | - Imran Shah
- Center for Computational Toxicology and Exposure, Office of Research and Development, U.S. Environmental Protection Agency, Research Triangle Park, North Carolina 27711, United States
| |
Collapse
|
2
|
Pierro JD, Ahir BK, Baker NC, Kleinstreuer NC, Xia M, Knudsen TB. Computational model for fetal skeletal defects potentially linked to disruption of retinoic acid signaling. Front Pharmacol 2022; 13:971296. [PMID: 36172177 PMCID: PMC9511990 DOI: 10.3389/fphar.2022.971296] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/16/2022] [Accepted: 07/21/2022] [Indexed: 11/13/2022] Open
Abstract
All-trans retinoic acid (ATRA) gradients determine skeletal patterning morphogenesis and can be disrupted by diverse genetic or environmental factors during pregnancy, leading to fetal skeleton defects. Adverse Outcome Pathway (AOP) frameworks for ATRA metabolism, signaling, and homeostasis allow for the development of new approach methods (NAMs) for predictive toxicology with less reliance on animal testing. Here, a data-driven model was constructed to identify chemicals associated with both ATRA pathway bioactivity and prenatal skeletal defects. The phenotype data was culled from ToxRefDB prenatal developmental toxicity studies and produced a list of 363 ToxRefDB chemicals with altered skeletal observations. Defects were classified regionally as cranial, post-cranial axial, appendicular, and other (unspecified) features based on ToxRefDB descriptors. To build a multivariate statistical model, high-throughput screening bioactivity data from >8,070 chemicals in ToxCast/Tox21 across 10 in vitro assays relevant to the retinoid signaling system were evaluated and compared to literature-based candidate reference chemicals in the dataset. There were 48 chemicals identified for effects on both in vivo skeletal defects and in vitro ATRA pathway targets for computational modeling. The list included 28 chemicals with prior evidence of skeletal defects linked to retinoid toxicity and 20 chemicals without prior evidence. The combination of thoracic cage defects and DR5 (direct repeats of 5 nucleotides for RAR/RXR transactivation) disruption was the most frequently occurring phenotypic and target disturbance, respectively. This data model provides valuable AOP elucidation and validates current mechanistic understanding. These findings also shed light on potential avenues for new mechanistic discoveries related to ATRA pathway disruption and associated skeletal dysmorphogenesis due to environmental exposures.
Collapse
Affiliation(s)
- Jocylin D. Pierro
- Center for Computational Toxicology and Exposure (CCTE), Computational Toxicology and Bioinformatics Branch (CTBB), Office of Research and Development (ORD), U.S. Environmental Protection Agency (USEPA), Research Triangle Park, NC, United States
| | - Bhavesh K. Ahir
- Eurofins Medical Device Testing, Lancaster, PA, United States
| | - Nancy C. Baker
- Scientific Computing and Data Curation Division (SCDCD), Leidos Contractor, Center for Computational Toxicology and Exposure (CCTE), USEPA/ORD, Research Triangle Park, NC, United States
| | - Nicole C. Kleinstreuer
- Interagency Center for the Evaluation of Alternative Toxicological Methods (NICEATM), National Toxicology Program, National Institutes of Health, Research Triangle Park, NC, United States
| | - Menghang Xia
- Division for Pre-Clinical Innovation, National Center for Advancing Translational Sciences, National Institutes of Health, Bethesda, MD, United States
| | - Thomas B. Knudsen
- Center for Computational Toxicology and Exposure (CCTE), Computational Toxicology and Bioinformatics Branch (CTBB), Office of Research and Development (ORD), U.S. Environmental Protection Agency (USEPA), Research Triangle Park, NC, United States
| |
Collapse
|