epistract

Search PubMed and download articles into a local corpus for epistract ingestion

Build knowledge graph from existing epistract extractions

Analyze epistemic patterns (hypotheses, contradictions, conditional claims) in a built knowledge graph and produce a Super Domain claims layer

Export knowledge graph to GraphML, GEXF, CSV, SQLite, or JSON

Ingest scientific documents and build a drug discovery knowledge graph

Search entities in the drug discovery knowledge graph

Install and verify epistract dependencies (sift-kg, optional RDKit/Biopython)

Validate molecular identifiers (SMILES, sequences) in extraction results

Open interactive knowledge graph visualization in browser

Fetch a batch of PubMed articles for the epistract corpus. Dispatched by /epistract-acquire when processing many articles in parallel. Each agent handles a batch of PMIDs independently.

Extract biomedical entities and relations from a single document for the epistract drug discovery knowledge graph. Dispatched by /epistract-ingest when processing multiple documents in parallel. Each agent handles one document independently.

Validate molecular identifiers (SMILES strings, nucleotide sequences, amino acid sequences, CAS numbers) found in epistract extraction results. Uses RDKit for chemistry and Biopython for sequences.

Use when extracting entities and relations from drug discovery, pharmaceutical, or biomedical documents. Activates for PubMed papers, bioRxiv preprints, clinical trial reports, FDA documents, patent filings, and any document discussing compounds, targets, mechanisms, diseases, or clinical trials.