René's URL Explorer Experiment


Title: RareFoldGPCR: Agonist Design Beyond Natural Amino Acids | bioRxiv

X Title: RareFoldGPCR: Agonist Design Beyond Natural Amino Acids

Description: bioRxiv - the preprint server for biology, operated by openRxiv, a nonprofit organization dedicated to advancing scientific communication

X Description: Noncanonical amino acids (NCAAs) expand the chemical diversity of peptides beyond the twenty standard residues, offering new opportunities for designing binders with novel interaction modes and functional activity. G protein-coupled receptors (GPCRs) are central to cellular signalling and represent one of the largest classes of therapeutic targets, yet their functional modulation remains challenging. Here, we present RareFoldGPCR (RFG), a GPCR-specialised AI model for structure prediction and design that supports NCAAs. By applying transfer learning on high-resolution GPCR structures, RFG accurately models and rationally designs both linear and cyclic peptides that incorporate NCAAs and modulate GPCR activity. This is achieved without the model ever being trained on NCAA-based GPCR modulators. We showcase the capability of RFG by designing peptide agonists for the glucagon-like peptide-1 receptor (GLP1R) and validating their functional activity experimentally in cell-based assays. We investigate the precise capabilities of generating active agonists by expanding different regions of the native GLP-1 hormone, and further demonstrate the design of cyclic peptide agonists with entirely novel sequences and topologies, creating new agonist modes. We analyse how design metrics relate to pathway specificity, enabling precise modulation of pathway activity, such as activating the cAMP response without recruiting β-arrestin to reduce receptor desensitisation. RFG shows how transfer learning on specific target classes enables generalisation to new chemistry and molecular topology, providing a broadly applicable strategy for designing functional ligands beyond the constraints of natural amino acid chemistry. RFG is freely available: ### Competing Interest Statement P.B. and T.H are co-founders of Cyclic Therapeutics, a company that designs cyclic peptides.

X: {at}biorxivpreprint

Generator: Drupal 7 (http://drupal.org)

direct link

Domain: www.biorxiv.org

Nonetext/html; charset=utf-8
google-site-verificationlH38pX61A6XH6RNu01u4WWWT9UnF9WhcST6isHEJv-I
article_thumbnailhttps://www.biorxiv.org/content/biorxiv/early/2025/10/03/2025.10.01.679733/embed/graphic-3.gif
typearticle
categoryarticle
HW.identifier/biorxiv/early/2025/10/03/2025.10.01.679733.atom
HW.pisabiorxiv;2025.10.01.679733v1
DC.Formattext/html
DC.Languageen
DC.TitleRareFoldGPCR: Agonist Design Beyond Natural Amino Acids
DC.Identifier10.1101/2025.10.01.679733
DC.Date2025-10-03
DC.PublisherCold Spring Harbor Laboratory
DC.Rights© 2025, Posted by Cold Spring Harbor Laboratory. This pre-print is available under a Creative Commons License (Attribution-NonCommercial 4.0 International), CC BY-NC 4.0, as described at http://creativecommons.org/licenses/by-nc/4.0/
DC.AccessRightsrestricted
DC.DescriptionNoncanonical amino acids (NCAAs) expand the chemical diversity of peptides beyond the twenty standard residues, offering new opportunities for designing binders with novel interaction modes and functional activity. G protein-coupled receptors (GPCRs) are central to cellular signalling and represent one of the largest classes of therapeutic targets, yet their functional modulation remains challenging. Here, we present RareFoldGPCR (RFG), a GPCR-specialised AI model for structure prediction and design that supports NCAAs. By applying transfer learning on high-resolution GPCR structures, RFG accurately models and rationally designs both linear and cyclic peptides that incorporate NCAAs and modulate GPCR activity. This is achieved without the model ever being trained on NCAA-based GPCR modulators. We showcase the capability of RFG by designing peptide agonists for the glucagon-like peptide-1 receptor (GLP1R) and validating their functional activity experimentally in cell-based assays. We investigate the precise capabilities of generating active agonists by expanding different regions of the native GLP-1 hormone, and further demonstrate the design of cyclic peptide agonists with entirely novel sequences and topologies, creating new agonist modes. We analyse how design metrics relate to pathway specificity, enabling precise modulation of pathway activity, such as activating the cAMP response without recruiting β-arrestin to reduce receptor desensitisation. RFG shows how transfer learning on specific target classes enables generalisation to new chemistry and molecular topology, providing a broadly applicable strategy for designing functional ligands beyond the constraints of natural amino acid chemistry. RFG is freely available: ### Competing Interest Statement P.B. and T.H are co-founders of Cyclic Therapeutics, a company that designs cyclic peptides.
DC.ContributorPatrick Bryant
article:published_time2025-10-03
article:sectionNew Results
citation_titleRareFoldGPCR: Agonist Design Beyond Natural Amino Acids
citation_abstract

Abstract

Noncanonical amino acids (NCAAs) expand the chemical diversity of peptides beyond the twenty standard residues, offering new opportunities for designing binders with novel interaction modes and functional activity. G protein-coupled receptors (GPCRs) are central to cellular signalling and represent one of the largest classes of therapeutic targets, yet their functional modulation remains challenging. Here, we present RareFoldGPCR (RFG), a GPCR-specialised AI model for structure prediction and design that supports NCAAs. By applying transfer learning on high-resolution GPCR structures, RFG accurately models and rationally designs both linear and cyclic peptides that incorporate NCAAs and modulate GPCR activity. This is achieved without the model ever being trained on NCAA-based GPCR modulators. We showcase the capability of RFG by designing peptide agonists for the glucagon-like peptide-1 receptor (GLP1R) and validating their functional activity experimentally in cell-based assays. We investigate the precise capabilities of generating active agonists by expanding different regions of the native GLP-1 hormone, and further demonstrate the design of cyclic peptide agonists with entirely novel sequences and topologies, creating new agonist modes. We analyse how design metrics relate to pathway specificity, enabling precise modulation of pathway activity, such as activating the cAMP response without recruiting β-arrestin to reduce receptor desensitisation. RFG shows how transfer learning on specific target classes enables generalisation to new chemistry and molecular topology, providing a broadly applicable strategy for designing functional ligands beyond the constraints of natural amino acid chemistry. RFG is freely available: https://github.com/patrickbryant1/RareFoldGPCR

citation_journal_titlebioRxiv
citation_publisherCold Spring Harbor Laboratory
citation_publication_date2025/01/01
citation_mjidbiorxiv;2025.10.01.679733v1
citation_id2025.10.01.679733v1
citation_public_urlhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1
citation_abstract_html_urlhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.abstract
citation_full_html_urlhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.full
citation_pdf_urlhttps://www.biorxiv.org/content/biorxiv/early/2025/10/03/2025.10.01.679733.full.pdf
citation_issn2692-8205
citation_doi10.1101/2025.10.01.679733
citation_num_pages32
citation_article_typeArticle
citation_sectionNew Results
citation_firstpage2025.10.01.679733
citation_authorPatrick Bryant
citation_author_institutionScience for Life Laboratory, The Department of Molecular Biosciences, The Wenner-Gren Institute, Stockholm University
citation_author_orcidhttp://orcid.org/0000-0003-3439-1866
citation_author_emailpatrick.bryant{at}scilifelab.se
citation_referenceSteinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nat Biotechnol. 2017;35: 1026–1028.
twitter:cardsummary
twitter:imagehttps://www.biorxiv.org/sites/default/files/images/biorxiv_logo_homepage7-5-small.png
og-titleRareFoldGPCR: Agonist Design Beyond Natural Amino Acids
og-urlhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1
og-site-namebioRxiv
og-descriptionNoncanonical amino acids (NCAAs) expand the chemical diversity of peptides beyond the twenty standard residues, offering new opportunities for designing binders with novel interaction modes and functional activity. G protein-coupled receptors (GPCRs) are central to cellular signalling and represent one of the largest classes of therapeutic targets, yet their functional modulation remains challenging. Here, we present RareFoldGPCR (RFG), a GPCR-specialised AI model for structure prediction and design that supports NCAAs. By applying transfer learning on high-resolution GPCR structures, RFG accurately models and rationally designs both linear and cyclic peptides that incorporate NCAAs and modulate GPCR activity. This is achieved without the model ever being trained on NCAA-based GPCR modulators. We showcase the capability of RFG by designing peptide agonists for the glucagon-like peptide-1 receptor (GLP1R) and validating their functional activity experimentally in cell-based assays. We investigate the precise capabilities of generating active agonists by expanding different regions of the native GLP-1 hormone, and further demonstrate the design of cyclic peptide agonists with entirely novel sequences and topologies, creating new agonist modes. We analyse how design metrics relate to pathway specificity, enabling precise modulation of pathway activity, such as activating the cAMP response without recruiting β-arrestin to reduce receptor desensitisation. RFG shows how transfer learning on specific target classes enables generalisation to new chemistry and molecular topology, providing a broadly applicable strategy for designing functional ligands beyond the constraints of natural amino acid chemistry. RFG is freely available: ### Competing Interest Statement P.B. and T.H are co-founders of Cyclic Therapeutics, a company that designs cyclic peptides.
og-typearticle
og-imagehttps://www.biorxiv.org/sites/default/files/images/biorxiv_logo_homepage7-5-small.png
citation_date2025-10-03

Links:

Skip to main contenthttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1#main-content
https://www.biorxiv.org/
Homehttps://www.biorxiv.org/
Abouthttps://www.biorxiv.org/about-biorxiv
Submithttps://www.biorxiv.org/submit-a-manuscript
ALERTS / RSShttps://www.biorxiv.org/alertsrss
Advanced Searchhttps://www.biorxiv.org/search
View ORCID Profilehttp://orcid.org/0009-0009-2913-4786
View ORCID Profilehttp://orcid.org/0000-0002-7384-092X
View ORCID Profilehttp://orcid.org/0000-0003-3439-1866
Find this author on Google Scholarhttps://www.biorxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Qiuzhen%2BLi%2B
Find this author on PubMedhttps://www.biorxiv.org/lookup/external-ref?access_num=Li%20Q&link_type=AUTHORSEARCH
Search for this author on this sitehttps://www.biorxiv.org/search/author1%3AQiuzhen%2BLi%2B
ORCID record for Qiuzhen Lihttp://orcid.org/0009-0009-2913-4786
Find this author on Google Scholarhttps://www.biorxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Thomas%2BHelleday%2B
Find this author on PubMedhttps://www.biorxiv.org/lookup/external-ref?access_num=Helleday%20T&link_type=AUTHORSEARCH
Search for this author on this sitehttps://www.biorxiv.org/search/author1%3AThomas%2BHelleday%2B
ORCID record for Thomas Helledayhttp://orcid.org/0000-0002-7384-092X
Find this author on Google Scholarhttps://www.biorxiv.org/lookup/google-scholar?link_type=googlescholar&gs_type=author&author%5B0%5D=Patrick%2BBryant%2B
Find this author on PubMedhttps://www.biorxiv.org/lookup/external-ref?access_num=Bryant%20P&link_type=AUTHORSEARCH
Search for this author on this sitehttps://www.biorxiv.org/search/author1%3APatrick%2BBryant%2B
ORCID record for Patrick Bryanthttp://orcid.org/0000-0003-3439-1866
patrick.bryant{at}scilifelab.sehttps://www.biorxiv.org/cdn-cgi/l/email-protection#c3b3a2b7b1aaa0a8eda1b1baa2adb7b8a2b7beb0a0aaafaaa5a6afa2a1edb0a6
Abstracthttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1
https://www.biorxiv.org/panels_ajax_tab/biorxiv_tab_art/node:4930450/1
Full Texthttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.full-text
https://www.biorxiv.org/panels_ajax_tab/article_tab_full_text/node:4930450/1
Info/Historyhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.article-info
https://www.biorxiv.org/panels_ajax_tab/biorxiv_tab_info/node:4930450/1
Metricshttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.article-metrics
https://www.biorxiv.org/panels_ajax_tab/article_tab_metrics/node:4930450/1
Data/Codehttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.external-links
https://www.biorxiv.org/panels_ajax_tab/article_tab_data_code/node:4930450/1
Preview PDFhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.full.pdf+html
https://www.biorxiv.org/panels_ajax_tab/biorxiv_tab_pdf/node:4930450/1
https://github.com/patrickbryant1/RareFoldGPCRhttps://github.com/patrickbryant1/RareFoldGPCR
https://zenodo.org/uploads/15180406https://zenodo.org/uploads/15180406
CC-BY-NC 4.0 International licensehttp://creativecommons.org/licenses/by-nc/4.0/
Back to tophttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1#page
Previoushttps://www.biorxiv.org/content/10.1101/2025.10.01.679747v1
Next https://www.biorxiv.org/content/10.1101/2025.10.01.677488v1
Download PDFhttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1.full.pdf
Data/Codehttps://www.biorxiv.org/content/early/2025/10/03/2025.10.01.679733.external-links
Emailhttps://www.biorxiv.org/
RareFoldGPCR: Agonist Design Beyond Natural Amino Acidshttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1
Sharehttps://www.biorxiv.org/
http://twitter.com/share?url=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&text=RareFoldGPCR%3A%20Agonist%20Design%20Beyond%20Natural%20Amino%20Acids
http://www.facebook.com/sharer.php?u=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&t=RareFoldGPCR%3A%20Agonist%20Design%20Beyond%20Natural%20Amino%20Acids
http://www.linkedin.com/shareArticle?mini=true&url=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&title=RareFoldGPCR%3A%20Agonist%20Design%20Beyond%20Natural%20Amino%20Acids&summary=&source=bioRxiv
http://www.mendeley.com/import/?url=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&title=RareFoldGPCR%3A%20Agonist%20Design%20Beyond%20Natural%20Amino%20Acids
Citation Toolshttps://www.biorxiv.org/
BibTeXhttps://www.biorxiv.org/highwire/citation/4930450/bibtext
Bookendshttps://www.biorxiv.org/highwire/citation/4930450/bookends
EasyBibhttps://www.biorxiv.org/highwire/citation/4930450/easybib
EndNote (tagged)https://www.biorxiv.org/highwire/citation/4930450/endnote-tagged
EndNote 8 (xml)https://www.biorxiv.org/highwire/citation/4930450/endnote-8-xml
Medlarshttps://www.biorxiv.org/highwire/citation/4930450/medlars
Mendeleyhttps://www.biorxiv.org/highwire/citation/4930450/mendeley
Papershttps://www.biorxiv.org/highwire/citation/4930450/papers
RefWorks Taggedhttps://www.biorxiv.org/highwire/citation/4930450/refworks-tagged
Ref Managerhttps://www.biorxiv.org/highwire/citation/4930450/reference-manager
RIShttps://www.biorxiv.org/highwire/citation/4930450/ris
Zoterohttps://www.biorxiv.org/highwire/citation/4930450/zotero
Tweet Widgethttp://twitter.com/share?url=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&count=horizontal&via=&text=RareFoldGPCR%3A%20Agonist%20Design%20Beyond%20Natural%20Amino%20Acids&counturl=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1
Facebook Likehttp://www.facebook.com/plugins/like.php?href=https%3A//www.biorxiv.org/content/10.1101/2025.10.01.679733v1&layout=button_count&show_faces=false&action=like&colorscheme=light&width=100&height=21&font=&locale=
Google Plus Onehttps://www.biorxiv.org/content/10.1101/2025.10.01.679733v1
Bioinformatics https://www.biorxiv.org/collection/bioinformatics
All Articleshttps://www.biorxiv.org/content/early/recent
Animal Behavior and Cognitionhttps://www.biorxiv.org/collection/animal-behavior-and-cognition
Biochemistryhttps://www.biorxiv.org/collection/biochemistry
Bioengineeringhttps://www.biorxiv.org/collection/bioengineering
Bioinformaticshttps://www.biorxiv.org/collection/bioinformatics
Biophysicshttps://www.biorxiv.org/collection/biophysics
Cancer Biologyhttps://www.biorxiv.org/collection/cancer-biology
Cell Biologyhttps://www.biorxiv.org/collection/cell-biology
Clinical Trialshttps://www.biorxiv.org/collection/clinical-trials
Developmental Biologyhttps://www.biorxiv.org/collection/developmental-biology
Ecologyhttps://www.biorxiv.org/collection/ecology
Epidemiologyhttps://www.biorxiv.org/collection/epidemiology
Evolutionary Biologyhttps://www.biorxiv.org/collection/evolutionary-biology
Geneticshttps://www.biorxiv.org/collection/genetics
Genomicshttps://www.biorxiv.org/collection/genomics
Immunologyhttps://www.biorxiv.org/collection/immunology
Microbiologyhttps://www.biorxiv.org/collection/microbiology
Molecular Biologyhttps://www.biorxiv.org/collection/molecular-biology
Neurosciencehttps://www.biorxiv.org/collection/neuroscience
Paleontologyhttps://www.biorxiv.org/collection/paleontology
Pathologyhttps://www.biorxiv.org/collection/pathology
Pharmacology and Toxicologyhttps://www.biorxiv.org/collection/pharmacology-and-toxicology
Physiologyhttps://www.biorxiv.org/collection/physiology
Plant Biologyhttps://www.biorxiv.org/collection/plant-biology
Scientific Communication and Educationhttps://www.biorxiv.org/collection/scientific-communication-and-education
Synthetic Biologyhttps://www.biorxiv.org/collection/synthetic-biology
Systems Biologyhttps://www.biorxiv.org/collection/systems-biology
Zoologyhttps://www.biorxiv.org/collection/zoology

Viewport: width=device-width, initial-scale=1


URLs of crawlers that visited me.