AMR-Evidence Extraction, Integration, and Interoperability
Extracting antimicrobial susceptibility and genotype–phenotype evidence from literature, harmonizing AMR and virulence databases, and combining that evidence with genome-based predictive models.
Goal
Extract antimicrobial susceptibility and genotype–phenotype evidence from literature, harmonize AMR and virulence databases, and combine literature-derived knowledge with genome-based predictive models.
Three-Day MVP
Use Salmonella and Staphylococcus aureus as demonstration organisms. Screen a defined PubMed/PMC corpus, extract AST measurements and genotype associations, normalize drug names and breakpoints, and link the evidence to BV-BRC genomes, AMRFinderPlus entities, CARD, VFDB, and related resources.
A small predictive component could combine literature-derived evidence with BV-BRC genomic features to predict resistance phenotype or rank plausible resistance mechanisms.
Evaluation
Manually curate 75–100 paper passages. Report extraction precision/recall, accession-linking accuracy, prediction F1/AUROC, provenance completeness, and disagreement with current AMR annotations.
Team
Team assignments are being finalized ahead of the codeathon. Participants can review project teams, and request reassignment, in the shared participant spreadsheet circulated by the organizing team.