[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-121518-en":3,"doc-seo-121518-105":30,"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":27,"seo_description":14,"update_tm":28,"read_time":29},121518,687197207919,"Theodora","https://ap-avatar.wpscdn.com/avatar/a000253d6f5f7c60be?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779446848396160552",8,"Research & Report","GLYCANML - A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning","Glycans are essential biomolecules whose rapidly growing functional datasets enable data-driven machine learning for glycan understanding. Yet, a standardized benchmark for glycan property and function prediction is still missing. GLYCANML fills this gap with a comprehensive suite covering glycan taxonomy, immunogenicity, glycosylation type, and protein-glycan interaction tasks. Using both tokenized sequences and graph representations, the benchmark evaluates sequence models and graph neural networks, and introduces GLYCANML-MTL to study multi-task transfer and performance gains, with public datasets, source code, and a leaderboard.","arXiv :2405 . 16206v 3 [ cs .LG] 1 Oct 2024  \nGLYCANML: A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning  \nMinghao Xu 1,2,3 Yunteng Geng 1 ∗ Yihang Zhang 1 ∗ Ling Yang 1 Jian Tang2,3,4,5 Wentao Zhang 1 †  \n*equal contribution †corresponding author  \n1Peking University 2Mila-Québec AI Institute 3BioGeometry  \n4HEC Montréal 5 CIFAR AI Research Chair  \n[contacts:](contacts: minghao.xu@mila.quebec)[ minghao.xu@mila.quebec](contacts: minghao.xu@mila.quebec), [wentao.zhang@pku.edu.cn](wentao.zhang@pku.edu.cn)  \nAbstract  \nGlycans are basic biomolecules and perform essential functions within living organisms. The rapid increase of functional glycan data provides a good opportunity for machine learning solutions to glycan understanding. However, there still lacks a standard machine learning benchmark for glycan property and function prediction.  \nIn this work, we fill this blank by building a comprehensive benchmark for Glycan Machine Learning (GLYCANML) . The GLYCANML benchmark consists of diverse types of tasks including glycan taxonomy prediction, glycan immunogenicity prediction, glycosylation type prediction, and protein-glycan interaction prediction.  \nGlycans can be represented by both sequences and graphs in GLYCANML, which enables us to extensively evaluate sequence-based models and graph neural networks (GNNs) on benchmark tasks. Furthermore, by concurrently performing eight glycan taxonomy prediction tasks, we introduce the GLYCANML-MTL testbed for multi-task learning (MTL) algorithms. Also, we evaluate how taxonomy prediction can boost other three function prediction tasks by MTL. Experimental results show the superiority of modeling glycans with multi-relational GNNs, and suitable MTL methods can further boost model performance. We provide all datasets and source codes at [https://github.com/GlycanML/GlycanML](https://github.com/GlycanML/GlycanML) and maintain  \na leaderboard at [https://GlycanML.github.io/project](https://GlycanML.github.io/project).  \n1 Introduction  \nGlycans are fundamental biomolecules that play crucial roles in maintaining the normal physiological functions and health status of living organisms. They can regulate inflammatory responses [24], enable the recognition and communication between cells [65], preserve stable blood sugar levels [6], etc. Thanks to the advance of high-throughput sequencing techniques of glycans [63, 32], a large number of glycan data are accessible, e.g., the more than 240 thousand glycans stored in the GlyTouCan database [50] . This progress enables glycan function analysis by machine learning methods which are essentially data-driven.  \nThere are some existing works that employ machine learning models to predict the species origins of glycans [11, 12], glycosylation phenomenon [40, 34] and the ability of glycans to induce immune response [58, 38] . These works mainly aim to solve one or several related glycan understanding problems. However, there still lacks a comprehensive benchmark studying the general effectiveness of various machine learning models on predicting diverse glycan properties and functions, which hinders the progress of machine learning for glycan understanding. As a matter of fact, comprehensive benchmark studies greatly facilitate the machine learning research of other biomolecules like small molecules [59, 51], proteins [44, 61] and nucleic acids [57, 39] .  \nPreprint. Under review.  \nTherefore, in this work, we take the initiative of building a Glycan Machine Learning (GLYCANML) benchmark featured with diverse types of tasks and multiple glycan representation structures. The GLYCANML benchmark consists of 11 benchmark tasks for understanding important glycan properties and functions, including glycan taxonomy prediction, glycan immunogenicity prediction, glycosylation type prediction, and protein-glycan interaction prediction. For each task, we carefully split the benchmark dataset to evaluate the generalization ability of machine l","cbCaiazkebkPAPNc","https://ap.wps.com/l/cbCaiazkebkPAPNc","pdf",1358804,1,14,"English","en",105,"# Introduction\n## Benchmark scope and tasks\n## Dataset splitting and generalization\n## Glycan representations and model evaluation\n## Multi-task learning testbed (GLYCANML-MTL)","[{\"question\":\"What is GLYCANML and what problems does it benchmark?\",\"answer\":\"GLYCANML is a comprehensive benchmark for Glycan Machine Learning that covers glycan taxonomy prediction, glycan immunogenicity prediction, glycosylation type prediction, and protein-glycan interaction prediction.\"},{\"question\":\"How does GLYCANML represent glycans for machine learning?\",\"answer\":\"It supports two representation structures: tokenized sequence representations and planar graph representations, enabling evaluation of both sequence-based models and graph neural networks.\"},{\"question\":\"What does the GLYCANML-MTL testbed evaluate?\",\"answer\":\"GLYCANML-MTL evaluates multi-task learning by jointly solving eight, correlated glycan taxonomy prediction tasks and measuring knowledge transfer across taxonomies to boost other function prediction tasks.\"}]","GLYCANML - A Multi-Task and Multi-Structure Benchmark for Glycan Machine Learning | PDF",1785736061,35,{"code":4,"msg":31,"data":32},"ok",{"site_id":24,"language":23,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"glycanml-a-multi-task-and-multi-structure-benchmark-for-glycan-machine-learning","",{"@graph":36,"@context":85},[37,54,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":53},"https://docshare.wps.com/document/glycanml-a-multi-task-and-multi-structure-benchmark-for-glycan-machine-learning/121518/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":62,"encodingFormat":61,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-03",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 is GLYCANML and what problems does it benchmark?","Question",{"text":75,"@type":76},"GLYCANML is a comprehensive benchmark for Glycan Machine Learning that covers glycan taxonomy prediction, glycan immunogenicity prediction, glycosylation type prediction, and protein-glycan interaction prediction.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does GLYCANML represent glycans for machine learning?",{"text":80,"@type":76},"It supports two representation structures: tokenized sequence representations and planar graph representations, enabling evaluation of both sequence-based models and graph neural networks.",{"name":82,"@type":73,"acceptedAnswer":83},"What does the GLYCANML-MTL testbed evaluate?",{"text":84,"@type":76},"GLYCANML-MTL evaluates multi-task learning by jointly solving eight, correlated glycan taxonomy prediction tasks and measuring knowledge transfer across taxonomies to boost other function prediction tasks.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":89,"canonical":52},"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":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]