[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85164-en":3,"doc-seo-85164-105":29,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},85164,962075114765,"Quinn","https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd",8,"Research & Report","Vilya-1: An all-atom foundation model for macrocycle structure prediction and design","Macrocyclic peptides are an increasingly important therapeutic modality, yet existing computational structure and property modeling methods cover only limited chemical scopes and generalize poorly across synthetically accessible space. Vilya-1 introduces a deep learning foundation model tackling two key issues: sampling biologically relevant conformations across arbitrary chemistries and predicting developability properties such as membrane permeability. Using a uniform all-atom representation and heterogeneous structural training data, Vilya-1 improves geometric accuracy over physics-based, co-folding, and conformer-generator baselines while supporting generative design of novel macrocycles with tailored structures and properties.","arXiv :2607 .09998v 1 [ cs .LG] 10 Jul 2026  \nVilya-1: An all-atom foundation model for macrocycle structure prediction and design  \nVilya Research  \nAbstract  \nMacrocyclic peptides are an increasingly important therapeutic modality, but existing computational methods for modeling their structures and properties are limited in scope and do not generalize well across the synthetically accessible chemical space. In this work, we introduce Vilya-1, a deep learning model that addresses two central challenges in macrocycle design: sampling biologically relevant conformations across arbitrary chemistries and predicting key developability properties such as membrane permeability. Vilya-1 operates on a uniform all-atom representation and is trained on heterogeneous structural datasets spanning diverse topologies and chemical classes. Across a broad set of macrocycles composed of canonical and non-canonical residues, Vilya-1 substantially improves geometric accuracy relative to physics-based methods, co-folding networks, and deep-learning conformer generators, while maintaining broad chemical coverage that extends to small molecules. Vilya-1 also supports generative applications, enabling the design of novel macrocycles with tailored chemical, structural, and property profiles. Together, these capabilities establish Vilya-1 as a foundation model for accelerating the development of next-generation macrocycle therapeutics.  \n1 Introduction  \nMacrocyclic peptide therapeutics offer the potential to bind to traditionally undruggable protein surfaces while retaining many of the pharmacological advantages of small molecules, particularly the ability to access intracellular compartments and the potential for oral dosing [1, 2] . De novo design studies have shown that macrocycles composed of canonical amino acids can be engineered with atomic level accuracy, vastly expanding the set of scaffolds beyond those found in nature [3, 4, 5] . More recently, computational pipelines have been extended to chemically diverse macrocyclic backbonesand non-peptidic macrocyclic oligoamides [6] . This chemical diversity opens the possibility for macrocycles to address challenging targets involving both intra and extracellular protein-protein interactions. These advances have highlighted the importance of accurately modeling the low-energy conformational states a macrocycle may adopt, as these states control target binding, permeability, and other drug-like properties. To date, however, conformational sampling has generally relied on custom computational procedures tailored to specific chemistries. In order to unlock the therapeutic potential of macrocycles, it is necessary to develop computational methods that generalize across synthetically accessible chemistriesand accurately capture biologically relevant low-energy states.  \nExisting methods to predict the structures of macrocyclic peptides largely derive from tools for protein structure prediction or small molecule conformer generation. Both have important limitations for chemically-diverse cyclic systems. Generalized structure prediction networks such as Boltz-2 and RosettaFold3 (RF3) [8, 9] can in principle model macrocycles, yet in practice do not generalize well to non-canonical chemistries and typically fail to produce sufficiently diverse conformational ensembles for macrocycles that may lack a single, well-defined ground state. Although there are examples of fine-tuning co-folding networks on structural data for macrocyclic peptides containing non-canonical amino acids, these approaches have not demonstrated robust generalization beyond the narrow chemistries represented in their training sets [10, 11, 12, 13] . Small-molecule conformer generators are designed for diversity but often do not generalize to larger, flexible molecules with ring constraints. Physics-based protocols that use molecular dynamics or stochastic sampling are capable of modeling highly complex macrocycles, but they are ineffici","cbCaijMJmqc7EQSg","https://ap.wps.com/l/cbCaijMJmqc7EQSg","pdf",7594873,1,21,"English","en",105,"# Introduction\n## Motivation for generalizable macrocycle modeling\n## Limits of existing structure prediction and conformer generation methods\n## Overview of Vilya-1 and its approach","[{\"question\":\"What problems does Vilya-1 address in macrocycle design?\",\"answer\":\"Vilya-1 targets two central challenges: sampling biologically relevant conformations across arbitrary chemistries and predicting developability properties such as membrane permeability.\"},{\"question\":\"How does Vilya-1 represent macrocycles for learning and prediction?\",\"answer\":\"Vilya-1 uses a uniform all-atom representation that does not distinguish between peptide and small-molecule chemistries, allowing one architecture to cover diverse macrocycle chemistries.\"},{\"question\":\"How does Vilya-1 compare with prior methods?\",\"answer\":\"The model substantially improves geometric accuracy relative to physics-based methods, co-folding networks, and deep-learning conformer generators, while maintaining broad chemical coverage and enabling generative macrocycle design.\"}]",1784201488,53,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":27},"vilya-1-an-all-atom-foundation-model-for-macrocycle-structure-prediction-and-design","",{"@graph":35,"@context":85},[36,53,68],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/vilya-1-an-all-atom-foundation-model-for-macrocycle-structure-prediction-and-design/85164/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problems does Vilya-1 address in macrocycle design?","Question",{"text":75,"@type":76},"Vilya-1 targets two central challenges: sampling biologically relevant conformations across arbitrary chemistries and predicting developability properties such as membrane permeability.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does Vilya-1 represent macrocycles for learning and prediction?",{"text":80,"@type":76},"Vilya-1 uses a uniform all-atom representation that does not distinguish between peptide and small-molecule chemistries, allowing one architecture to cover diverse macrocycle chemistries.",{"name":82,"@type":73,"acceptedAnswer":83},"How does Vilya-1 compare with prior methods?",{"text":84,"@type":76},"The model substantially improves geometric accuracy relative to physics-based methods, co-folding networks, and deep-learning conformer generators, while maintaining broad chemical coverage and enabling generative macrocycle design.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":45,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":45,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":45,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":45,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":45,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]