Skip to content
Protean LabsDocs

Evidence Retrieval

The evidence index is a local semantic layer over Protean’s scientific context. It retrieves relevant records before candidate explanations, candidate papers, failure-aware analysis, and comparative review.

The index currently reads from:

  • processed evidence records
  • literature mention records
  • failure records
  • generated assessment context when available
  • patent or raw records when present in the local data layer

Index Contract

The index stores text, metadata, and embeddings in a local SQLite file.

data/cache/evidence_index.sqlite
-> documents
-> text embeddings
-> source metadata
-> retrieval scores

Sequence-like queries are rejected by the evidence index. Peptide and protein sequence comparison belongs to the ESM sequence layer, not the text retrieval layer.

Retrieval Use

Retrieved evidence is used to improve context selection. It helps papers and explanations speak from the best available local scientific material instead of dumping raw runtime fields.

It does not:

  • establish assay validation
  • establish efficacy
  • establish safety
  • determine patentability
  • replace deterministic candidate validators

Scientific Posture

Evidence retrieval is a context layer. It helps humans and local reasoning models focus attention, but the downstream reality check remains wet-lab validation and founder/scientific review.

PreviousText EmbeddingsNextReranking