--- name: signaling-kb-batched description: Extract all entity types (interventions, pathways, biomarkers, conflicts) from Google Drive transcript analyses using a three-stage LLM pipeline with Claude Opus 4.5 for extraction and Haiku 4.5 for intelligent merging. allowed-tools: Read, Edit, Create, Glob, Grep, Execute, TodoWrite user_invocable: true argument-hint: [--from-tracker] [--status] [--dry-run] --- # Signaling KB Batched - Full Entity Extraction Pipeline Extract **all entity types** from transcript analyses with: - **4 entity types**: Interventions, Pathways, Biomarkers, Conflicts - **LLM-based semantic deduplication** (Opus 4.5) - resolves aliases like "Sirolimus" = "Rapamycin" - **LLM-based intelligent merging** (Haiku 4.5) - merges new data into existing entities - **Google Drive sync** with processing tracker - **Source-agnostic output** (no document titles) - **PMID/DOI citations only** (invalid citations filtered) - **Resume capability** via tracker ## Architecture ``` ┌─────────────────────────────────────────────────────────────────────────────┐ │ THREE-STAGE LLM PIPELINE │ │ │ │ ┌──────────────────┐ ┌─────────────────────┐ ┌──────────────────────┐ │ │ │ EXTRACTION │──▶│ SEMANTIC DEDUP │──▶│ LLM MERGE │ │ │ │ Claude Opus 4.5 │ │ Claude Opus 4.5 │ │ Claude Haiku 4.5 │ │ │ │ (64k tokens) │ │ (64k tokens) │ │ (64k tokens) │ │ │ └──────────────────┘ └─────────────────────┘ └──────────────────────┘ │ │ │ │ Stage 1: Extract all Stage 2: Map extracted Stage 3: Intelligently │ │ entity types from entities to existing merge new data into │ │ transcript KB entries (aliases) existing JSON files │ └─────────────────────────────────────────────────────────────────────────────┘ ``` ## Quick Start ```bash # 1. Sync transcripts from Google Drive (with tracking) python3 scripts/sync_gdrive_transcripts.py \ --folder-id 1ajmDczDiX1hbMsSAkyG2Lat6SmCN8Q5T \ --recursive # 2. Check status python3 scripts/extract_kb_openrouter.py --status # 3. Run extraction (resumes from tracker) OPENROUTER_API_KEY=xxx python3 scripts/extract_kb_openrouter.py --from-tracker # 4. For long-running extraction, use background mode nohup python3 scripts/extract_kb_openrouter.py --from-tracker > extraction.log 2>&1 & ``` ## Entity Types Extracted ### 1. Interventions Compounds, peptides, biologics, modalities, and lifestyle interventions. ```json { "intervention_id": "rapamycin", "aliases": ["sirolimus", "Rapamune"], "category": "small_molecule", "dosing_protocols": [{ "indication": "longevity/healthspan", "dose": "5-6mg", "frequency": "weekly", "route": "oral", "peer_reviewed_citations": [{"pmid": "24566879", "title": "..."}], "confidence": "moderate" }], "mechanism_claims": [{ "target": "MTORC1", "effect": "inhibit", "peer_reviewed_citations": [], "confidence": "high" }], "contraindications": ["active infection", "wound healing"], "monitoring": ["lipid panel", "CBC", "fasting glucose"] } ``` ### 2. Pathways Signaling cascades mapped to SIGNAL-10 axes with crosstalk relationships. ```json { "pathway_id": "mtor-signaling", "pathway_name": "mTOR signaling", "axis": 1, "axis_name": "Nutrient Sensing & Energetics", "description": "Master regulator of cell growth...", "core_nodes": ["MTOR", "RPTOR", "RICTOR", "EIF4EBP1"], "upstream_triggers": ["amino acids", "insulin", "IGF-1"], "downstream_effects": ["protein synthesis", "autophagy inhibition"], "interventions_targeting": [ {"intervention_name": "rapamycin", "effect": "inhibit"} ], "crosstalk": [ {"target_pathway": "autophagy", "relationship": "inhibits"} ], "confidence": "high" } ``` ### 3. Biomarkers Lab values with interpretation patterns and context-specific optimal ranges. ```json { "biomarker_id": "igf-1", "biomarker_name": "IGF-1", "full_name": "Insulin-like Growth Factor 1", "reference_range": {"low": 100, "high": 300, "unit": "ng/mL"}, "interpretation_patterns": [{ "pattern_id": "low-igf1-with-insulin-resistance", "biomarker_state": {"igf1": "low", "homa_ir": "elevated"}, "interpretation": "Suggests GH resistance or hepatic dysfunction...", "differential": ["hepatic_insulin_resistance", "gh_resistance"], "clinical_implications": "Address insulin resistance before GH therapy", "confidence": "moderate" }], "optimal_ranges_by_context": [{ "context": "longevity", "optimal": {"low": 150, "high": 250}, "rationale": "Below 150 suggests GH deficiency..." }], "interventions_affecting": [ {"intervention_name": "growth hormone", "effect": "increase"} ], "pathways_involved": ["igf1-axis"] } ``` ### 4. Conflicts Contradictions between claims requiring clinical judgment. ```json { "conflict_id": "conflict_rapa_pgc1a_001", "conflict_type": "mechanism_dispute", "entity_type": "intervention", "entity_name": "rapamycin", "claim_a": "Rapamycin inhibits PGC-1alpha", "claim_b": "Rapamycin has biphasic effect - acute inhibition but chronic upregulation", "context": "Different timeframes may explain apparent contradiction", "resolution_status": "unresolved" } ``` ## LLM Pipeline Details ### Stage 1: Extraction (Opus 4.5) - Extracts all 4 entity types from transcript - Parses embedded references (PMIDs/DOIs from References section) - Assigns confidence levels based on evidence quality - Cost: ~$1.50 per transcript ### Stage 2: Semantic Entity Resolution (Opus 4.5) - Maps extracted entity names to existing KB entries - Identifies aliases (e.g., "Sirolimus" → "rapamycin") - Prevents duplicate entries for same entity - Example mappings: - "NAD+" = "Nicotinamide adenine dinucleotide" - "GLP-1 agonists" = "Glucagon-like peptide-1 agonists" - "Vitamin D" = "Cholecalciferol" = "D3" ### Stage 3: LLM Merge (Haiku 4.5) - Intelligently merges new data into existing entities - Preserves existing information (never deletes) - Deduplicates dosing protocols and mechanism claims - Handles conflicting information by keeping both positions - Cost: ~$0.10 per entity merge ## Cost Tracking ### API Pricing (January 2026 via OpenRouter) | Model | Input | Output | |-------|-------|--------| | Claude Opus 4.5 | $5.00/M tokens | $25.00/M tokens | | Claude Haiku 4.5 | $0.80/M tokens | $4.00/M tokens | ### Typical Costs | Operation | Cost | |-----------|------| | Single transcript (all entities) | ~$1.50-2.00 | | Full batch (224 transcripts) | ~$250-300 | | LLM merge per entity | ~$0.05-0.15 | ### Configuration All operations use **64,000 max_tokens** for maximum output capacity: - Extraction: 64k tokens - Entity resolution: 64k tokens - LLM merge: 64k tokens ## Source-Agnostic Output **CRITICAL**: Output contains NO source-identifying information. **Included:** - Claims and mechanisms (rephrased, not verbatim) - Citations with valid PMIDs or DOIs only - Source type classification (`"source_type": "expert_lecture"`) - Confidence levels **Excluded:** - Document titles or names - Course/institute names - Speaker names - Any metadata that could bias downstream models ## Key Files | File | Purpose | |------|---------| | `scripts/extract_kb_openrouter.py` | Main extraction script with 3-stage pipeline | | `scripts/sync_gdrive_transcripts.py` | Google Drive sync with tracking | | `scripts/monitor_extraction.sh` | Auto-restart monitor for long runs | | `.signaling-kb/state/processing-tracker.json` | Tracks processed vs pending | | `.signaling-kb/interventions/*.json` | Intervention entity nodes | | `.signaling-kb/pathways/*.json` | Pathway entity nodes | | `.signaling-kb/biomarkers/*.json` | Biomarker entity nodes | | `.signaling-kb/conflicts/*.json` | Conflict entity nodes | ## Monitoring Long Runs For extraction runs that may take hours: ```bash # Start the monitor (checks every 10 minutes, auto-restarts if stopped) nohup bash scripts/monitor_extraction.sh > monitor.log 2>&1 & # Check monitor status tail -f monitor.log # Check extraction progress python3 scripts/extract_kb_openrouter.py --status ``` ## Changelog | Version | Date | Changes | |---------|------|---------| | 1.0.0 | 2026-01-29 | Full 4-entity extraction with LLM merge and semantic dedup | | 0.3.0 | 2026-01-23 | Added Google Drive sync, tracking, source-agnostic output | | 0.2.0 | 2026-01-23 | Upgraded to Claude Opus 4.5 | | 0.1.0 | 2026-01-23 | Initial implementation (interventions only) |