{"schema_version":"1.0","hypothesis_id":39642,"triple":{"drug":"pioglitazone","drug_canonical":"pioglitazone","mediator":"GFAP","mediator_canonical":"GFAP","outcome":"Alzheimer's disease","outcome_canonical":"Alzheimer disease"},"domain":{"domain":"cross_domain","source":"type_2_diabetes","target":"alzheimer"},"provenance_completeness":"partial_backfill","completeness_notes":"Backfilled retroactively from DB state on 2026-05-13. The pipeline_run this row references was created by the backfill script, not by the actual generation orchestrator. Source papers, claims, scoring, and embedding config are recovered; generation-time metadata (exact git commit at gen time, host, exact command) is not recoverable.","source_papers":[{"doi":"10.1371/journal.pone.0291995","pmid":"38236817","year":2024,"roles":["mediator_outcome"],"title":"Large-scale gene expression changes in APP/PSEN1 and GFAP mutation models exhibit high congruence with Alzheimer’s disease","journal":"PLoS One","work_id":68030,"publication_date":"2024-01-18"},{"doi":"10.3233/jad-170799","pmid":"30040709","year":2018,"roles":["mediator_outcome"],"title":"A Selection of Important Genes and Their Correlated Behavior in Alzheimer’s Disease","journal":"J Alzheimers Dis","work_id":69627,"publication_date":"2018-07-20"},{"doi":"10.3233/jad-170205","pmid":"30400085","year":2018,"roles":["mediator_outcome"],"title":"Changes in Expression Profiles Revealed by Transcriptomic Analysis in Peripheral Blood Mononuclear Cells of Alzheimer’s Disease Patients","journal":"J Alzheimers Dis","work_id":61640,"publication_date":"2018-11-02"},{"doi":"10.1097/01.j.pain.0000460333.79127.be","pmid":"25599238","year":2015,"roles":["drug_mediator"],"title":"Pioglitazone rapidly reduces neuropathic pain through astrocyte and nongenomic PPARγ mechanisms","journal":"Pain","work_id":242781,"publication_date":"2015-01-14"}],"source_paper_count":4,"extracted_claims":[{"side":"drug_mediator","object":"GFAP","context":"in spinal cord dorsal horn after nerve injury","subject":"Pioglitazone","work_id":242781,"claim_id":189584,"polarity":"positive","predicate":"downregulates","confidence":0.85,"extraction_run_id":27},{"side":"mediator_outcome","object":"Alzheimer's disease","context":"in peripheral blood mononuclear cells of AD patients","subject":"GFAP","work_id":61640,"claim_id":40792,"polarity":"positive","predicate":"associated_with","confidence":0.7,"extraction_run_id":16},{"side":"mediator_outcome","object":"Alzheimer's disease","context":"in CNS gene expression analysis of animal models","subject":"GFAP","work_id":68030,"claim_id":55441,"polarity":"positive","predicate":"associated_with","confidence":0.9,"extraction_run_id":16},{"side":"mediator_outcome","object":"Alzheimer's disease","context":"in AD neurons with neurofibrillary tangles","subject":"GFAP","work_id":69627,"claim_id":58539,"polarity":"positive","predicate":"associated_with","confidence":0.6,"extraction_run_id":16}],"extracted_claim_count":4,"extraction":{"model":"deepseek-v4-flash","model_source":"configured","prompt_version":"v1_2026_05_07","run_ids":[16,27]},"bridge":{"a_to_b_predicate":"downregulates","b_to_c_predicate":"associated_with","a_to_b_evidence_count":1,"b_to_c_evidence_count":3,"edge_support_log_ab":0.6931471805599453,"edge_support_log_bc":1.3862943611198906,"novelty_factor":1.0,"direct_a_to_c_count":0},"embedding":{"model":"bge-m3","dim":1024,"max_length":8192,"similarity_scores":{}},"scoring":{"method":"swanson_abc_log_evidence_with_novelty_penalty","formula":"score = min((log(ab+1) + log(bc+1)) * novelty / 10, 1.0); novelty = 1.0 if direct==0 else 0.7 / (direct+1)","score_raw":0.20794415416798356,"score":0.20794415416798356,"quality_tier":"mid","quality_reason":null},"pipeline_run":{"id":4,"run_uuid":"eb8b7730-d92e-4cd5-95dd-95f6c5d62cc2","command":"scripts/backfill_provenance_v2.py:v2_2026_05_12","status":"completed","git_commit":"9996ffaa56e9b7cd5f677bd8203624c2c18aa000","host":"MacBook-Pro-Daniel96.local","started_at":"2026-05-12T21:35:53.363326+00:00","ended_at":"2026-05-12T21:43:41.903387+00:00","notes":"Retroactive provenance backfill on 2026-05-13 for snapshot v2_2026_05_12. Synthetic pipeline_runs row — original generation run records are lost (pre-date provenance schema). All recoverable fields populated from current DB state (claims, works, repurposing_hypotheses); pipeline_run record is synthetic, hence completeness='partial_backfill'.\nBackfilled 1,618 provenance rows. Source papers: avg=3.73, zero=0. Claims matched: avg=4.09, zero=0.\nRefreshed 1,618 provenance rows. Source papers: avg=3.73, zero=0. Claims matched: avg=4.09, zero=0.\nRefreshed 1,618 provenance rows. Source papers: avg=3.73, zero=0. Claims matched: avg=4.09, zero=0.","domains":["als","alzheimer","depression","epilepsy","glioblastoma","inflammatory_bowel_disease","multiple_sclerosis","pancreatic_cancer","rheumatoid_arthritis","type_2_diabetes"]},"generated_at":"2026-05-12T21:43:40.945575+00:00"}