Metraipolis › archive › dc_58eb3af27fba
Is Retrieval All You Need? Assessment and Emergence of Novelty in Protein Structure Generation — arXiv:2608.10598
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.10598
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T06:47:01.345Z
- SHA-256 of the archived text
6e97caf21202289b55c4e085952ac2efd27633ba061606b8edcc571023d4706f- Stable URL of this page
https://metraipolis.com/documento/dc_58eb3af27fba
How to cite
Is Retrieval All You Need? Assessment and Emergence of Novelty in Protein Structure Generation — arXiv:2608.10598 — arXiv (en), https://arxiv.org/abs/2608.10598 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_58eb3af27fba (Metraipolis archive dc_58eb3af27fba, sha256 6e97caf21202289b55c4e085952ac2efd27633ba061606b8edcc571023d4706f)
BibTeX
@misc{dc58eb3af27fba,
title = {Is Retrieval All You Need? Assessment and Emergence of Novelty in Protein Structure Generation — arXiv:2608.10598},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.10598},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_58eb3af27fba; sha256 6e97caf21202289b55c4e085952ac2efd27633ba061606b8edcc571023d4706f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.10598. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Protein backbone generation models are often credited with exploring novel fold space based solely on low full-chain similarity to known proteins, yet this cannot distinguish a genuinely new fold from a novel assembly of known structural units. We first ask whether this granularity mismatch alone explains the reported rates, and introduce the Domain Retrieval Rate (DRR), the fraction of generated backbones for which any constituent domain matches a known domain in CATH S40. Applied to eight backbone generation models spanning diffusion and flow-matching paradigms, DRR finds locally alignable known structure in most outputs, while the fraction containing a substantially covered complete domain is considerably smaller and depends on the scoring convention. To calibrate what retrieval alone can achieve, we propose RetFold, a zero-training baseline that constructs backbones by retrieving CATH domains and refining inter-domain connections through geometry-based helix-linker optimization, at two orders of magnitude lower cost on CPU alone. Tema degli esploratori: scienze della vita e salute (retro-catalogo arXiv: annunci del 2026-08-12 in q-bio.BM). Categorie arXiv: Biomolecules (q-bio.BM). Pagina del lavoro: https://arxiv.org/abs/2608.10598 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-19T06:40:34.608Z dalla pagina https://arxiv.org/catchup/q-bio.BM/2026-08-12?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).