[Atlas] Causal KG entity resolution — bridge 19K free-text causal edges to canonical KG entities open

SciDEX has two parallel KG structures: causal_edges (19,753 rows, free-text source_entity/target_entity with mechanism_description and evidence_pmids) vs kg_edges (2,366 rows, typed entity references). These are NOT connected. The 19K causal edges are the richest mechanistic knowledge in SciDEX — genes to pathways to diseases, with evidence PMIDs — but they are not queryable via KG API, not surfaced in hypothesis scoring, and not linked to canonical entities. What to do: 1. Build entity resolution pipeline: map causal_edges text entities to canonical_entities using fuzzy + semantic matching 2. For resolved pairs, create corresponding kg_edges entries 3. Track resolution quality (confidence, unresolved fraction) 4. Enrich hypothesis KG-connectivity scores with new edges Match strategy: exact lowercase -> alias lookup -> substring -> LLM fuzzy Confidence tiers: exact=1.0, alias=0.95, fuzzy>=0.8=record with confidence, fuzzy<0.8=skip Read first: docs/planning/specs/quest_atlas_causal_kg_entity_resolution.md Success criteria per iteration: >= 5,000 causal edges entity-resolved and linked to kg_edges.

Last Error

acquire_fail:worktree_creation_failed:branch_held_by_other:held_by=/tmp/cer-iter5

Git Commits (4)

{
  "completion_shas": [
    "2ccd91247f1a6c136ed48d8e1a43091a93fd14c4",
    "c1792a8e81272363b897f9fe4b3e7d506167171f"
  ],
  "completion_shas_checked_at": ""
}

Sibling Tasks in Quest (Atlas) ↗