{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:GUMD27NWB4SOQ43GDQXV3PVUZ6","short_pith_number":"pith:GUMD27NW","schema_version":"1.0","canonical_sha256":"35183d7db60f24e873661c2f5dbeb4cf97004623978a8db34e5e9bb42a50aee9","source":{"kind":"arxiv","id":"2605.03690","version":2},"attestation_state":"computed","paper":{"title":"Graph Neural Network based Hierarchy-Aware Embeddings of Knowledge Graphs: Applications to Yeast Phenotype Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Graph neural networks enriched with semantic loss produce hierarchy-aware embeddings of yeast knowledge graphs that predict cell growth phenotypes from gene knockouts.","cross_cats":["cs.AI","q-bio.QM"],"primary_cat":"cs.LG","authors_text":"Alexander H. Gower, Daniel Brunns{\\aa}ker, Filip Kronstr\\\"om, Ievgeniia A. Tiukova, Ross D. King","submitted_at":"2026-05-05T12:34:45Z","abstract_excerpt":"We present a method for finding hierarchy-aware embeddings of knowledge graphs (KGs) using graph neural networks (GNNs) enriched with a semantic loss derived from underlying ontologies. This method yields embeddings that better reflect domain knowledge. To demonstrate their utility, we predict and interpret the effects of gene deletions in the yeast Saccharomyces cerevisiae and learn box embeddings for KGs in the absence of a prediction task. We further show how box embeddings can serve as the basis for evaluating KG revisions.\n  Our yeast KG is constructed from community databases and ontolog"},"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":"2605.03690","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-05-05T12:34:45Z","cross_cats_sorted":["cs.AI","q-bio.QM"],"title_canon_sha256":"da590924b7f95036a43a1b252a5ce1f1c08c54665566fbeb78145d272bb03369","abstract_canon_sha256":"ddaaf7316591cd842e042a7696e886b089bae987d79678d3139c19be5aa5ebf7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-21T02:05:03.888911Z","signature_b64":"+pmryP7jPD7juJ9V3CI1qPgdrux+EF+UbhCoWHB+63e/pihi1VgeJQqPs+9uLXY9xIRfMWbeLSqChSEPMjapDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"35183d7db60f24e873661c2f5dbeb4cf97004623978a8db34e5e9bb42a50aee9","last_reissued_at":"2026-05-21T02:05:03.888277Z","signature_status":"signed_v1","first_computed_at":"2026-05-21T02:05:03.888277Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Graph Neural Network based Hierarchy-Aware Embeddings of Knowledge Graphs: Applications to Yeast Phenotype Prediction","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Graph neural networks enriched with semantic loss produce hierarchy-aware embeddings of yeast knowledge graphs that predict cell growth phenotypes from gene knockouts.","cross_cats":["cs.AI","q-bio.QM"],"primary_cat":"cs.LG","authors_text":"Alexander H. Gower, Daniel Brunns{\\aa}ker, Filip Kronstr\\\"om, Ievgeniia A. Tiukova, Ross D. King","submitted_at":"2026-05-05T12:34:45Z","abstract_excerpt":"We present a method for finding hierarchy-aware embeddings of knowledge graphs (KGs) using graph neural networks (GNNs) enriched with a semantic loss derived from underlying ontologies. This method yields embeddings that better reflect domain knowledge. To demonstrate their utility, we predict and interpret the effects of gene deletions in the yeast Saccharomyces cerevisiae and learn box embeddings for KGs in the absence of a prediction task. We further show how box embeddings can serve as the basis for evaluating KG revisions.\n  Our yeast KG is constructed from community databases and ontolog"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Over 10-fold cross validation, these predictions have a mean R² score of 0.360, significantly higher than baseline comparisons... Incorporating semantic loss terms in the training of the models improves their predictive performance (R²=0.377) by aligning embeddings with ontology structure.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the yeast knowledge graph constructed from community databases and ontology terms contains the biologically relevant relationships needed to predict cell-growth phenotypes, and that the semantic loss term correctly enforces hierarchy without introducing new biases or data leakage.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"GNNs with ontology-derived semantic loss create hierarchy-aware KG embeddings that predict yeast double gene knockout phenotypes with mean R²=0.360 (improved to 0.377 with semantic loss), outperforming baselines, generalizing to triple knockouts, and supporting experimental hypothesis validation.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Graph neural networks enriched with semantic loss produce hierarchy-aware embeddings of yeast knowledge graphs that predict cell growth phenotypes from gene knockouts.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"5abdf1a961f7d32a6f1c0c10dfd6acf5ac1cd1b6383c7b2c19e4134600b0a697"},"source":{"id":"2605.03690","kind":"arxiv","version":2},"verdict":{"id":"5f936277-320b-4c39-b0ae-92dff2de6591","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T16:58:24.121907Z","strongest_claim":"Over 10-fold cross validation, these predictions have a mean R² score of 0.360, significantly higher than baseline comparisons... Incorporating semantic loss terms in the training of the models improves their predictive performance (R²=0.377) by aligning embeddings with ontology structure.","one_line_summary":"GNNs with ontology-derived semantic loss create hierarchy-aware KG embeddings that predict yeast double gene knockout phenotypes with mean R²=0.360 (improved to 0.377 with semantic loss), outperforming baselines, generalizing to triple knockouts, and supporting experimental hypothesis validation.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the yeast knowledge graph constructed from community databases and ontology terms contains the biologically relevant relationships needed to predict cell-growth phenotypes, and that the semantic loss term correctly enforces hierarchy without introducing new biases or data leakage.","pith_extraction_headline":"Graph neural networks enriched with semantic loss produce hierarchy-aware embeddings of yeast knowledge graphs that predict cell growth phenotypes from gene knockouts."},"integrity":{"clean":false,"summary":{"advisory":1,"critical":0,"by_detector":{"doi_compliance":{"total":1,"advisory":1,"critical":0,"informational":0}},"informational":0},"endpoint":"/pith/2605.03690/integrity.json","findings":[{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1101/2025.06.24.661378v3.ISSN) was visible in the surrounding text but could not be confirmed against doi.org as printed.","detector":"doi_compliance","severity":"advisory","ref_index":5,"audited_at":"2026-05-19T15:05:40.036948Z","detected_doi":"10.1101/2025.06.24.661378v3.ISSN","finding_type":"recoverable_identifier","verdict_class":"incontrovertible","detected_arxiv_id":null}],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T13:36:19.512514Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_title_agreement","ran_at":"2026-05-20T00:31:21.498213Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T15:05:40.036948Z","status":"completed","version":"1.0.0","findings_count":1}],"snapshot_sha256":"29a80e154ea4ad7b1c029502e63fb9b3058c853e89d164fe7babaef91597f52c"},"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":"2605.03690","created_at":"2026-05-21T02:05:03.888380+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.03690v2","created_at":"2026-05-21T02:05:03.888380+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.03690","created_at":"2026-05-21T02:05:03.888380+00:00"},{"alias_kind":"pith_short_12","alias_value":"GUMD27NWB4SO","created_at":"2026-05-21T02:05:03.888380+00:00"},{"alias_kind":"pith_short_16","alias_value":"GUMD27NWB4SOQ43G","created_at":"2026-05-21T02:05:03.888380+00:00"},{"alias_kind":"pith_short_8","alias_value":"GUMD27NW","created_at":"2026-05-21T02:05:03.888380+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6","json":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6.json","graph_json":"https://pith.science/api/pith-number/GUMD27NWB4SOQ43GDQXV3PVUZ6/graph.json","events_json":"https://pith.science/api/pith-number/GUMD27NWB4SOQ43GDQXV3PVUZ6/events.json","paper":"https://pith.science/paper/GUMD27NW"},"agent_actions":{"view_html":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6","download_json":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6.json","view_paper":"https://pith.science/paper/GUMD27NW","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.03690&json=true","fetch_graph":"https://pith.science/api/pith-number/GUMD27NWB4SOQ43GDQXV3PVUZ6/graph.json","fetch_events":"https://pith.science/api/pith-number/GUMD27NWB4SOQ43GDQXV3PVUZ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6/action/storage_attestation","attest_author":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6/action/author_attestation","sign_citation":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6/action/citation_signature","submit_replication":"https://pith.science/pith/GUMD27NWB4SOQ43GDQXV3PVUZ6/action/replication_record"}},"created_at":"2026-05-21T02:05:03.888380+00:00","updated_at":"2026-05-21T02:05:03.888380+00:00"}