Hypothesis #40847

resveratrol → SIRT1 → tumor-initiating capacity

Drug domain: Type 2 diabetes Outcome domain: Pancreatic cancer Score: 0.312 Quality: high

← Back to all hypotheses

Provenance trace

Audit trail · resveratrol → SIRT1 → tumor-initiating capacity

4 source papers, 4 extracted claims, scored by Swanson ABC chain.

Backfilled

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.

1Source papers4
Filter
2Extracted claims4

Each row is one structured triple extracted by deepseek-v4-flash (prompt v1_2026_05_07, source: configured). Confidence is the LLM's self-rating.

  • resveratrolupregulatesSIRT1A→B
    "in PBMCs of T2DM patients with CHD"confidence 0.90· extraction run #27· claim #166800
  • ResveratrolupregulatesSIRT1A→B
    "in GK rat hearts"confidence 0.80· extraction run #27· claim #190273
  • ResveratrolactivatesSIRT1A→B
    "in diabetic male rats"confidence 0.70· extraction run #27· claim #199146
  • SIRT1increasestumor-initiating capacityB→C
    "in vivo mouse model (NOD/SCID mice)"confidence 0.90· extraction run #17· claim #64501
3Bridging logicSwanson ABC
Drug
resveratrol
activates
→
3 papers
Mediator
SIRT1
increases
→
1 paper
Outcome
tumor-initiating capacity
edge_support_log_ab
log(ab + 1) = 1.386
edge_support_log_bc
log(bc + 1) = 0.693
novelty_factor
1.000 (no direct drug→outcome evidence — full novelty)
direct_a_to_c_count
0
4Embedding modelbge-m3

Embeddings are used at the L1 filtering stage to prune off-topic abstracts before claim extraction. They do not directly score the hypothesis.

model
bge-m3
dim
1024
max_length
8192 tokens
similarity_scores
No per-hypothesis cosine scores stored.
5Final scoringhigh · 0.312
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)
raw score
0.2079
outreach score
0.3119
quality tier
high
Pipeline version
9996ffa
Pipeline run
id 4 · uuid eb8b7730…
Pipeline command
scripts/backfill_provenance_v2.py:v2_2026_05_12
Pipeline status
completed
Generated
2026-05-12 21:43:41 UTC
Schema version
1.0
Download full provenance JSON

Horizontal exploration

Similar hypotheses

Closest 8 of 1,618 hypotheses, ranked by hybrid score (50% bge-m3 embedding cosine · 40% drug/mediator/outcome overlap · 10% shared evidence papers).

  1. New indication0.94
    resveratrol→SIRT1→stemness marker expression

    Same drug + same mediator, different disease

    Pancreatic cancerhigh
  2. New indication0.91
    resveratrol→SIRT1→proliferation

    Same drug + same mediator, different disease

    Pancreatic cancerhigh
  3. New indication0.89
    resveratrol→SIRT1→patient survival

    Same drug + same mediator, different disease

    Pancreatic cancerhigh
  4. New indication0.88
    resveratrol→SIRT1→TNF-alpha secretion

    Same drug + same mediator, different disease

    Rheumatoid arthritishigh
  5. New indication0.88
    resveratrol→SIRT1→neuronal cell death

    Same drug + same mediator, different disease

    Epilepsyhigh
Show 3 more
  1. New indication0.88
    resveratrol→SIRT1→overexpression

    Same drug + same mediator, different disease

    ALShigh
  2. New indication0.88
    resveratrol→SIRT1→optokinetic response

    Same drug + same mediator, different disease

    Multiple sclerosishigh
  3. New indication0.88
    resveratrol→SIRT1→axon demyelination

    Same drug + same mediator, different disease

    Multiple sclerosishigh
Score trajectory

How this hypothesis evolved

First captured May 12, 2026, latest May 13, 2026. 2 snapshots.

0.000.100.200.300.4001234 score evidence count May 12, 2026 · score 0.312 · 4 evidence · CT: none · Lit: partial_evidenceMay 12freezeMay 13, 2026 · score 0.312 · 4 evidence · CT: none · Lit: partial_evidenceMay 13monthly
Score (left axis) Evidence count (right axis, dashed)