{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:SVU57ISD2GHSBEL2SHEMDNXDWZ","short_pith_number":"pith:SVU57ISD","schema_version":"1.0","canonical_sha256":"9569dfa243d18f20917a91c8c1b6e3b649dba4c55bbeb26436a4bb804c6789c1","source":{"kind":"arxiv","id":"2506.03028","version":1},"attestation_state":"computed","paper":{"title":"Protein Inverse Folding From Structure Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.BM"],"primary_cat":"cs.LG","authors_text":"Guangyong Chen, Jie Hu, Jiezhong Qiu, Junde Xu, Le Song, Pheng-Ann Heng, Xingyi Cheng, Xinyi Zhou, Zijun Gao","submitted_at":"2025-06-03T16:02:12Z","abstract_excerpt":"The inverse folding problem, aiming to design amino acid sequences that fold into desired three-dimensional structures, is pivotal for various biotechnological applications. Here, we introduce a novel approach leveraging Direct Preference Optimization (DPO) to fine-tune an inverse folding model using feedback from a protein folding model. Given a target protein structure, we begin by sampling candidate sequences from the inverse-folding model, then predict the three-dimensional structure of each sequence with the folding model to generate pairwise structural-preference labels. These labels are"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2506.03028","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-03T16:02:12Z","cross_cats_sorted":["q-bio.BM"],"title_canon_sha256":"a668953ca645e635c2a0b56a74c8b489c63a106f0ff62b137d0825f7ac4990c2","abstract_canon_sha256":"d068253b150f77bbca88c3398bbacc4ad1d2953d94fe4e0b8f34c9826955632e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:15:13.582626Z","signature_b64":"BW9T4FiS6PI98ClDzA+YYxhBOi0oaiGfy9EmC4/v47iH7HI82RDWB56HdbXBzAYpKDbY3em1WUlQJmuYIfbbCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9569dfa243d18f20917a91c8c1b6e3b649dba4c55bbeb26436a4bb804c6789c1","last_reissued_at":"2026-07-05T11:15:13.582125Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:15:13.582125Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Protein Inverse Folding From Structure Feedback","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["q-bio.BM"],"primary_cat":"cs.LG","authors_text":"Guangyong Chen, Jie Hu, Jiezhong Qiu, Junde Xu, Le Song, Pheng-Ann Heng, Xingyi Cheng, Xinyi Zhou, Zijun Gao","submitted_at":"2025-06-03T16:02:12Z","abstract_excerpt":"The inverse folding problem, aiming to design amino acid sequences that fold into desired three-dimensional structures, is pivotal for various biotechnological applications. Here, we introduce a novel approach leveraging Direct Preference Optimization (DPO) to fine-tune an inverse folding model using feedback from a protein folding model. Given a target protein structure, we begin by sampling candidate sequences from the inverse-folding model, then predict the three-dimensional structure of each sequence with the folding model to generate pairwise structural-preference labels. These labels are"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.03028","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2506.03028/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2506.03028","created_at":"2026-07-05T11:15:13.582186+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.03028v1","created_at":"2026-07-05T11:15:13.582186+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.03028","created_at":"2026-07-05T11:15:13.582186+00:00"},{"alias_kind":"pith_short_12","alias_value":"SVU57ISD2GHS","created_at":"2026-07-05T11:15:13.582186+00:00"},{"alias_kind":"pith_short_16","alias_value":"SVU57ISD2GHSBEL2","created_at":"2026-07-05T11:15:13.582186+00:00"},{"alias_kind":"pith_short_8","alias_value":"SVU57ISD","created_at":"2026-07-05T11:15:13.582186+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07567","citing_title":"SurfDesign: Effective Protein Design on Molecular Surfaces","ref_index":87,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23407","citing_title":"GeoCycler: Reward-Aligned 3D Diffusion for Constraint-Conditioned Cyclic Peptide Design","ref_index":80,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ","json":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ.json","graph_json":"https://pith.science/api/pith-number/SVU57ISD2GHSBEL2SHEMDNXDWZ/graph.json","events_json":"https://pith.science/api/pith-number/SVU57ISD2GHSBEL2SHEMDNXDWZ/events.json","paper":"https://pith.science/paper/SVU57ISD"},"agent_actions":{"view_html":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ","download_json":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ.json","view_paper":"https://pith.science/paper/SVU57ISD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.03028&json=true","fetch_graph":"https://pith.science/api/pith-number/SVU57ISD2GHSBEL2SHEMDNXDWZ/graph.json","fetch_events":"https://pith.science/api/pith-number/SVU57ISD2GHSBEL2SHEMDNXDWZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ/action/storage_attestation","attest_author":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ/action/author_attestation","sign_citation":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ/action/citation_signature","submit_replication":"https://pith.science/pith/SVU57ISD2GHSBEL2SHEMDNXDWZ/action/replication_record"}},"created_at":"2026-07-05T11:15:13.582186+00:00","updated_at":"2026-07-05T11:15:13.582186+00:00"}