{"data":{"event":{"id":"e9851137-de46-42ec-8789-8da57a4b21ad","slug":"an-interpretable-peptide-hla-model-emergently-learns-binding-energetics--9944ea1834","title":"\n\nAn interpretable peptide-HLA model emergently learns binding energetics and structure \n\n","short_summary":"The range of peptides a human leukocyte antigen (HLA) binds and displays modulates immune response and therefore underpins vaccine design, neoantigen discovery, autoimmunity, transplantation, and hypersensitivity reactions. Modern predictors of peptide-HLA (pMHC) binding and presentation are remarkably accurate, but they are black boxes; their internal computations are opaque and post-hoc explanatory methods lack guarantees of attribution. Here we introduce LAtent Motif INteraction Aggregation (LAMINA), an architecture whose prediction is, by construction, interpretable and attributable. LAMIN","full_description":"The range of peptides a human leukocyte antigen (HLA) binds and displays modulates immune response and therefore underpins vaccine design, neoantigen discovery, autoimmunity, transplantation, and hypersensitivity reactions. Modern predictors of peptide-HLA (pMHC) binding and presentation are remarkably accurate, but they are black boxes; their internal computations are opaque and post-hoc explanatory methods lack guarantees of attribution. Here we introduce LAtent Motif INteraction Aggregation (LAMINA), an architecture whose prediction is, by construction, interpretable and attributable. LAMINA embeds every gapless sub-sequence of the HLA pseudosequence and of the candidate peptide as a learned \"soft\" motif, scores every HLA-peptide-motif pair, and lastly aggregates those scores to produce a prediction. Despite having only 4.7 million parameters and training in about ten hours on a single desktop workstation, LAMINA matches or exceeds state-of-the-art predictors on held-out binding-affinity regression and is competitive on rank correlation. Strikingly, the model states correlate with interaction energies and other structural metrics in structurally characterized pMHC complexes, despite training exclusively on sequence data alone.\n\nUpdate (2026-03-11): Interpreting Black Box Reward Models. ARGO distills black-box reward models into interpretable rubrics using reinforcement learning.","ledger_type":"benefit","primary_domain_id":"fc7d617f-c9df-4b9f-b595-4408915ea380","event_status":"provisional","event_date":"2026-09-14T00:00:00.000Z","discovery_date":"2026-09-14T00:00:00.000Z","first_published_date":"2026-09-14T00:00:00.000Z","last_reviewed_date":"2026-03-11T00:00:00.000Z","geographic_scope":"International","affected_population":null,"base_impact_tier":1,"base_score":"1.00","attribution_multiplier":"0.1000","evidence_multiplier":"0.2000","realization_multiplier":"0.2000","durability_multiplier":"0.5000","current_event_score":"0.002000","confidence_level":"low","score_explanation":"Ingest update: independent source raised evidence; added a follow-up source. Score remains provisional until review.","methodology_version_id":"8745541b-0f8e-4d5a-9b47-5291d3813a8d","original_methodology_version_id":"8745541b-0f8e-4d5a-9b47-5291d3813a8d","published_at":"2026-09-15T04:00:52.237Z","created_at":"2026-09-15T04:00:52.237Z","updated_at":"2026-09-17T03:01:08.826Z","flags":[],"domain_name":"Other","domain_slug":"other","methodology_version":"0.2"},"contributions":[{"id":"afadfb6b-cc21-420c-a882-907aa20e8a84","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","model_id":"535014f7-c941-4d39-aa9b-8f9c1d3d07b7","role_description":"Unspecified system mentioned or implied by a news item. Remap to a named release when identified.","attribution_multiplier":"0.1000","credit_share":"1.0000","contribution_score":"0.002000","attribution_rationale":"News ingest does not infer a named model from the publisher alone. Attribution stays unspecified until a release is identified.","attribution_confidence":"medium","first_used_date":"2026-09-14T00:00:00.000Z","model_version_if_known":null,"review_status":"approved","created_at":"2026-09-15T04:00:52.279Z","model_slug":"unspecified-ai-system","model_name":"Unspecified AI system","identity_class":"unknown","is_internal":false,"family_name":"Unspecified","family_slug":"unknown-unspecified","organization_name":"Unknown","organization_slug":"unknown"}],"sources":[{"id":"aa2126f1-118d-4605-8255-b45dd4049363","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","url":"\nhttps://www.biorxiv.org/content/10.64898/2026.09.08.750216v1?rss=1\n","canonical_url":"\nhttps://www.biorxiv.org/content/10.64898/2026.09.08.750216v1?rss=1\n","source_type":"preprint","publisher":"bioRxiv","author":null,"publication_date":"2026-09-14T00:00:00.000Z","retrieved_at":"2026-09-15T04:00:52.333Z","title":"\n\nAn interpretable peptide-HLA model emergently learns binding energetics and structure \n\n","excerpt":"The range of peptides a human leukocyte antigen (HLA) binds and displays modulates immune response and therefore underpins vaccine design, neoantigen discovery, autoimmunity, transplantation, and hypersensitivity reactions. Modern predictors of peptide-HLA (pMHC) binding and presentation are remarkably accurate, but they are black boxes; their internal computations are opaque and post-hoc explanatory methods lack guarantees of attribution. Here we introduce LAtent Motif INteraction Aggregation (","content_hash":null,"source_reliability_class":"medium","is_primary_source":true,"is_independent":true,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-15T04:00:52.333Z"},{"id":"5a4b06a4-bcc6-4c73-9c82-aa581589b360","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","url":"https://alignment.openai.com/argo/","canonical_url":"https://alignment.openai.com/argo/","source_type":"developer_announcement","publisher":"OpenAI Alignment","author":null,"publication_date":"2026-03-11T00:00:00.000Z","retrieved_at":"2026-09-17T03:01:08.814Z","title":"Interpreting Black Box Reward Models","excerpt":"ARGO distills black-box reward models into interpretable rubrics using reinforcement learning.","content_hash":null,"source_reliability_class":"medium","is_primary_source":false,"is_independent":true,"is_peer_reviewed":false,"archived_url":null,"created_at":"2026-09-17T03:01:08.814Z"}],"claims":[{"id":"de7824f3-dfa7-4e57-a94e-1c7bc1a24e3a","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","claim_text":"\n\nAn interpretable peptide-HLA model emergently learns binding energetics and structure \n\n","claim_type":"outcome","claim_status":"supported","confidence_score":null,"created_at":"2026-09-15T04:00:52.345Z","updated_at":"2026-09-15T04:00:52.345Z"}],"revisions":[{"id":"0e75b0e2-d83e-4200-8b44-fb8609de5593","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","model_id":null,"previous_score":"0.000000","new_score":"0.001000","previous_factors":{},"new_factors":{"evidence":0.1,"base_score":1,"durability":0.5,"attribution":0.1,"realization":0.2},"change_reason":"Auto-published from news ingest.","trigger_type":"news_ingest","trigger_source_ids":null,"reviewer_id":null,"review_status":"published","created_at":"2026-09-15T04:00:52.362Z"},{"id":"63a4eb61-67c6-43f5-be0c-b4de0cdaf852","event_id":"e9851137-de46-42ec-8789-8da57a4b21ad","model_id":null,"previous_score":"0.001000","new_score":"0.002000","previous_factors":{"tier":1,"evidence":0.1,"base_score":1,"durability":0.5,"attribution":0.1,"realization":0.2},"new_factors":{"tier":1,"models":["535014f7-c941-4d39-aa9b-8f9c1d3d07b7"],"evidence":0.2,"base_score":1,"durability":0.5,"attribution":0.1,"realization":0.2},"change_reason":"Ingest update: independent source raised evidence; added a follow-up source.","trigger_type":"ingest_update","trigger_source_ids":["5a4b06a4-bcc6-4c73-9c82-aa581589b360"],"reviewer_id":null,"review_status":"published","created_at":"2026-09-17T03:01:08.833Z"}],"secondary":[]},"methodology_version":"0.1"}