The abstract is read at the publisher; the summary is JClub's.
Pharmacolibrary v26.11 links PK, PD, and PGx evidence to source-traceable executable models, enabling browser-based simulations across Modelica, FMI, SBML, MATLAB, and CellML.
- Why it matters: Published pharmacological knowledge is fragmented across literature and tools, which hinders systematic reuse, comparison, and reproducibility of computational models derived from that evidence.
- What they did: A multi-agent workflow uses LLM-assisted retrieval and extraction, then deterministic validation, ontology mapping, provenance tracking, automated model generation, and rule-based review to produce executable models.
- The result: Preliminary evaluation shows multi-stage extraction improves parameter recovery versus single-pass extraction, and locally deployed open-weight models achieve performance at least comparable to frontier commercial LLMs.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.