Katalog Plus
Bibliothek der Frankfurt UAS
Bald neuer Katalog: sichern Sie sich schon vorab Ihre persönlichen Merklisten im Nutzerkonto: Anleitung.
Dieses Ergebnis aus BASE kann Gästen nicht angezeigt werden.  Login für vollen Zugriff.

Protriever:End-to-End Differentiable Protein Homology Search for Fitness Prediction

Title: Protriever:End-to-End Differentiable Protein Homology Search for Fitness Prediction
Authors: Weitzman, Ruben; Groth, Peter Mørch; Van Niekerk, Lood; Otani, Aoi; Gal, Yarin; Marks, Debora S.; Notin, Pascal
Source: Weitzman , R , Groth , P M , Van Niekerk , L , Otani , A , Gal , Y , Marks , D S & Notin , P 2025 , Protriever : End-to-End Differentiable Protein Homology Search for Fitness Prediction . in Proceedings of the 42nd International Conference on Machine Learnin . PMLR , Proceedings of Machine Learning Research , vol. 267 , pp. 66397-66418 , 42nd International Conference on Machine Learning, ICML 2025 , Vancouver , Canada , 13/07/2025 . < https://proceedings.mlr.press/v267/ >
Publisher Information: PMLR
Publication Year: 2025
Collection: University of Copenhagen: Research / Forskning ved Københavns Universitet
Description: Retrieving homologous protein sequences is essential for a broad range of protein modeling tasks such as fitness prediction, protein design, structure modeling, and protein-protein interactions. Traditional workflows have relied on a two-step process: first retrieving homologs via Multiple Sequence Alignments (MSA), then training models on one or more of these alignments. However, MSA-based retrieval is computationally expensive, struggles with highly divergent sequences or complex insertions & deletions patterns, and operates independently of the downstream modeling objective. We introduce Protriever, an end-to-end differentiable framework that learns to retrieve relevant homologs while simultaneously training for the target task. When applied to protein fitness prediction, Protriever achieves state-of-the-art performance compared to sequence-based models that rely on MSA-based homolog retrieval, while being two orders of magnitude faster through efficient vector search. Protriever is both architectureand task-agnostic, and can flexibly adapt to different retrieval strategies and protein databases at inference time – offering a scalable alternative to alignment-centric approaches.
Document Type: article in journal/newspaper
File Description: application/pdf
Language: English
Availability: https://researchprofiles.ku.dk/da/publications/e5758ced-919b-4a91-bde1-ed762effd903; https://curis.ku.dk/ws/files/525589548/weitzman25a.pdf; https://proceedings.mlr.press/v267/
Rights: info:eu-repo/semantics/openAccess
Accession Number: edsbas.73C544CC
Database: BASE