{"data":{"event":{"id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","slug":"alphafold-3-biomolecular-complexes","title":"AlphaFold 3 predicts structures of proteins bound to DNA, RNA, and ligands","short_summary":"Google DeepMind and Isomorphic Labs published AlphaFold 3 in Nature, extending structure prediction from proteins to complexes that include nucleic acids, small molecules, ions, and modified residues.","full_description":"On 8 May 2024, Nature published AlphaFold 3, a diffusion-based model that predicts the joint structure of biomolecular complexes across nearly all molecular types in the Protein Data Bank. The paper reported higher accuracy than specialized docking and nucleic-acid tools on several interaction classes, including protein–ligand and antibody–antigen cases relative to AlphaFold-Multimer v2.3. Google later released academic-use code and weights and a public AlphaFold Server. The result is a widely used research tool rather than an approved clinical product; several later open models (Chai-1, Boltz-1) independently targeted the same task.","ledger_type":"benefit","primary_domain_id":"8f1af1b9-7302-4b51-aae9-fdfac04d158a","event_status":"verified","event_date":"2024-05-08T00:00:00.000Z","discovery_date":"2024-05-08T00:00:00.000Z","first_published_date":"2024-05-08T00:00:00.000Z","last_reviewed_date":"2024-05-08T00:00:00.000Z","geographic_scope":"Global research community","affected_population":"Structural biologists and drug-discovery groups using the model or server","base_impact_tier":4,"base_score":"30.00","attribution_multiplier":"0.9500","evidence_multiplier":"0.7000","realization_multiplier":"0.6000","durability_multiplier":"0.9000","current_event_score":"10.773000","confidence_level":"high","score_explanation":"Tier 4 reflects a field-level advance with a peer-reviewed method and public research access, not a civilizational health outcome. Attribution is high because the model is the result. Evidence is peer-review level (0.70), not independent wet-lab replication of every claim. Realization is experimental/tool deployment, not approved therapeutics. Durability is high for a reusable method.","methodology_version_id":"d7881163-fb23-4e71-8625-bac1a8662c0f","original_methodology_version_id":"d7881163-fb23-4e71-8625-bac1a8662c0f","published_at":"2024-05-08T00:00:00.000Z","created_at":"2026-09-13T20:47:20.075Z","updated_at":"2026-09-13T20:47:20.075Z","flags":[],"domain_name":"Biology","domain_slug":"biology","methodology_version":"0.1"},"contributions":[{"id":"0599f374-0585-4448-91a1-ca5f9a28468e","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","model_id":"ec34d05c-b6ad-4ce8-a1c1-95fb5ed4896e","role_description":"Produced the structure-prediction model jointly with Isomorphic Labs.","attribution_multiplier":"0.9500","credit_share":"1.0000","contribution_score":"10.773000","attribution_rationale":"The published architecture and trained model are the central scientific artifact.","attribution_confidence":"medium","first_used_date":"2024-05-08T00:00:00.000Z","model_version_if_known":null,"review_status":"approved","created_at":"2026-09-13T20:47:20.075Z","model_slug":"alphafold-3","model_name":"AlphaFold 3","identity_class":"named_release","is_internal":false,"family_name":"AlphaFold","family_slug":"google-deepmind-alphafold","organization_name":"Google DeepMind","organization_slug":"google-deepmind"}],"sources":[{"id":"4140b6f6-b895-4fa6-8aa1-df295de75022","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","url":"https://www.nature.com/articles/s41586-024-07487-w","canonical_url":"https://www.nature.com/articles/s41586-024-07487-w","source_type":"peer_reviewed","publisher":"Nature","author":null,"publication_date":"2024-05-08T00:00:00.000Z","retrieved_at":"2026-09-13T20:47:20.075Z","title":"Accurate structure prediction of biomolecular interactions with AlphaFold 3","excerpt":null,"content_hash":null,"source_reliability_class":"high","is_primary_source":true,"is_independent":true,"is_peer_reviewed":true,"archived_url":null,"created_at":"2026-09-13T20:47:20.075Z"},{"id":"64fcbe42-b86e-44d0-8e36-a009cfd5df32","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","url":"https://blog.google/technology/ai/google-deepmind-isomorphic-alphafold-3-ai-model/","canonical_url":"https://blog.google/technology/ai/google-deepmind-isomorphic-alphafold-3-ai-model/","source_type":"developer_announcement","publisher":"Google","author":null,"publication_date":"2024-05-08T00:00:00.000Z","retrieved_at":"2026-09-13T20:47:20.075Z","title":"Google DeepMind and Isomorphic Labs introduce AlphaFold 3","excerpt":null,"content_hash":null,"source_reliability_class":"low","is_primary_source":true,"is_independent":false,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-13T20:47:20.075Z"}],"claims":[{"id":"14f0e0dc-b303-43fa-b0b7-d96366c73f57","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","claim_text":"AlphaFold 3 predicts joint structures of proteins with nucleic acids, ligands, ions, and modified residues in a single model.","claim_type":"outcome","claim_status":"supported","confidence_score":null,"created_at":"2026-09-13T20:47:20.075Z","updated_at":"2026-09-13T20:47:20.075Z"},{"id":"c5ef0ceb-463c-453c-a9d8-d39cf0ab68e0","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","claim_text":"The model improved accuracy over specialized tools on several interaction classes reported in the Nature paper.","claim_type":"significance","claim_status":"supported","confidence_score":null,"created_at":"2026-09-13T20:47:20.075Z","updated_at":"2026-09-13T20:47:20.075Z"},{"id":"1b4a125e-3d66-4243-b6c2-e062865eb550","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","claim_text":"AlphaFold 3 has produced approved medicines.","claim_type":"outcome","claim_status":"disputed","confidence_score":null,"created_at":"2026-09-13T20:47:20.075Z","updated_at":"2026-09-13T20:47:20.075Z"}],"revisions":[{"id":"3d134894-ce79-44fe-9044-0cfbb45cc451","event_id":"505fdb6a-f11f-49d8-ae98-05fb45785bcd","model_id":null,"previous_score":"0.000000","new_score":"10.773000","previous_factors":{},"new_factors":{"evidence":0.7,"base_score":30,"durability":0.9,"attribution":0.95,"realization":0.6},"change_reason":"Initial adjudicated seed score under methodology 0.1.","trigger_type":"seed","trigger_source_ids":null,"reviewer_id":null,"review_status":"published","created_at":"2026-09-13T20:47:20.075Z"}],"secondary":[{"name":"Medicine & Health","slug":"medicine-health"},{"name":"Chemistry","slug":"chemistry"}]},"methodology_version":"0.1"}