[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-416514-105":59,"doc-detail-416514-en":122},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":115,"head_meta":117,"extra_data":119,"updated_unix":121},105,"en","deepadr-multimodal-prediction-of-adverse-drug-reaction-frequency","DeepADR - Multimodal prediction of adverse drug reaction frequency","","Adverse drug reactions (ADRs) drive major clinical trial failures and postmarket withdrawals, creating substantial risks to public health and delaying drug development. While computational alternatives reduce reliance on expensive preclinical tests, many existing models underperform on new compounds because they require prior drug–ADR knowledge or insufficiently integrate heterogeneous data. DeepADR presents a multimodal deep learning framework that predicts both ADR occurrence and ADR frequency using early-stage, readily available drug discovery information. It combines chemical structures, biological target profiles, and semantic ADR representations from a large language model. These modalities are fused through a Kolmogorov–Arnold Network to model nonlinear cross-modal relationships. Experiments show improved performance over prior methods on both classification and regression tasks, with robust generalization to new entities. By integrating chemical, biological, and semantic datasets, DeepADR supports early safety assessment and candidate prioritization and suggests a scalable approach with broader toxicological prediction potential.",{"@graph":69,"@context":114},[70,84,105],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":35,"@type":76,"position":81},"https://docshare.wps.com/document/healthcare/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/deepadr-multimodal-prediction-of-adverse-drug-reaction-frequency/416514/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/deepadr-multimodal-prediction-of-adverse-drug-reaction-frequency/416514.png","ImageObject",300,407,{"name":92,"@type":93},"Franzy","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-02","2026-09-28",true,{"@type":102,"interactionType":103,"userInteractionCount":44},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108],{"name":109,"@type":110,"acceptedAnswer":111},"Why is DeepADR expected to generalize to new chemical entities?","Question",{"text":112,"@type":113},"DeepADR is designed to leverage early-stage readily available multimodal drug discovery data, combining chemical, biological, and semantic information. The reported results show robust generalization to new entities, improving reliability beyond compound-specific prior knowledge.","Answer","https://schema.org",{"og:url":83,"og:type":116,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":118,"canonical":83},"index,follow",{"doc_id":120,"site_id":62},416514,1791292041,{"code":4,"msg":5,"data":123},{"doc_id":120,"user_id":124,"nickname":92,"user_avatar":125,"doc_module":4,"category_id":34,"category_name":35,"doc_title":65,"doc_description":67,"doc_content":126,"file_id":127,"file_url":128,"file_type":129,"file_size":130,"view_count":44,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":131,"language":132,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":133,"faqs":134,"seo_title":135,"seo_description":67,"update_tm":136,"read_time":137,"preview_page_count":4},2336478945519,"https://ap-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45","Briefings in Bioinformatics, 2025, 26(6), bbaf695  \n[https://doi.org/10.1093/bib/bbaf695](https://doi.org/10.1093/bib/bbaf695)  \nProblem Solving Protocol  \nDeepADR: multimodal prediction of adverse drug reaction frequency by integrating early-stage drug discovery information via Kolmogorov–Arnold networks  \nJingting Wan1 , 2 , 3 , Chenyang Jia1 , Danhong Dong1 , 2 , 3 , Yigang Chen1 , 2 , 3 , Yang-Chi-Dung Lin1 , 2 , 3 , Yisheng He1 , *, Hsi-Yuan Huang1 , 2 , 3 , *, Hsien-Da Huang1 , 2 , 3 ,4 , *  \n1 School of Medicine, The Chinese University of Hong Kong, Shenzhen, No. 2001 Longxiang Road, Longgang District, Shenzhen, Guangdong 518172, P.R. China  \n2Warshel Institute for Computational Biology, School of Medicine, The Chinese University of Hong Kong, Shenzhen, No. 2001 Longxiang Road, Longgang District, Shenzhen, Guangdong 518172, P.R. China  \n3 Guangdong Provincial Key Laboratory of Digital Biology and Drug Development, The Chinese University of Hong Kong, Shenzhen, No. 2001 Longxiang Road , Longgang District, Shenzhen, Guangdong 518172, P.R. China  \n4 Department of Endocrinology, Key Laboratory of Endocrinology of National Health Commission, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences & Peking Union Medical College, No. 157 Malianwa North Road, Haidian District, Beijing 100730, P.R. China  \n*Corresponding authors. Yisheng He, School of Medicine, The Chinese University of Hong Kong, Shenzhen, Guangdong 518172, P.R. China. [E-mail:](E-mail: heyisheng@cuhk.edu.cn)[ heyisheng@cuhk.edu.cn](E-mail: heyisheng@cuhk.edu.cn); Hsi-Yuan Huang, School of Medicine, The Chinese University of Hong Kong, Shenzhen, Guangdong 518172, P.R. China. [E-mail:](E-mail: huanghsiyuan@cuhk.edu.cn)[ huanghsiyuan@cuhk.edu.cn](E-mail: huanghsiyuan@cuhk.edu.cn); Hsien-Da Huang, School of Medicine, The Chinese University of Hong Kong, Shenzhen, Guangdong 518172, P.R. China. [E-mail:](E-mail: huanghsienda@cuhk.edu.cn)[ huanghsienda@cuhk.edu.cn](E-mail: huanghsienda@cuhk.edu.cn).  \nAbstract  \nAdverse drug reactions (ADRs) are a major cause of clinical trial failure and postmarket withdrawal, posing significant risks to public health and impeding drug development. While computational methods offer an alternative to costly preclinical testing, existing models often fail with novel compounds by requiring pre-existing information such as drug-ADR associations or by inadequately integrating diverse data sources. Here, we introduce DeepADR, a multimodal deep learning framework for predicting both the occurrence and frequency of ADRs using early-stage, readily available data. DeepADR integrates chemical structures and biological target profiles with semantic representations ofADR terms derived from a large language model (LLMs) . These heterogeneous parameters are fused using a Kolmogorov–Arnold Network (KAN), which enhances the modeling of complex, nonlinear relationships among modalities to improve predictive performance. Our model outperforms existing methods in predicting both ADR occurrence and frequency, demonstrating robust generalization to new chemical entities. DeepADR showed consistently better performance than other models across both classification and regression tasks. By effectively integrating chemical, biological, and semantic datasets, DeepADR provides a powerful, scalable tool for the early-stage safety assessment and candidate prioritization. This framework not only facilitates the prioritization of safer drug candidates but also offers a methodology for predicting the toxicity of other hazardous materials, holding significant promise for advancing public health.  \nKeywords: adverse drug reactions; drug development; multi modal deep learning; molecular representation; Kolmogorov–Arnold network  \nIntroduction  \nEnsuring drug safety throughout the pharmaceutical development pipeline is critical, as adverse drug reactions (ADRs) remain a substantial challenge. They contribute significantly to late-stage clinical trial ","cbCaitAVIjgAI22k","https://ap.wps.com/l/cbCaitAVIjgAI22k","pdf",1633394,13,"English","# Introduction\n## Motivation and challenges in ADR prediction\n## Need for computational methods\n## Existing computational approaches and scope\n# DeepADR framework and approach\n## Multimodal inputs from early-stage drug discovery\n## Semantic ADR representations from LLMs\n## Kolmogorov–Arnold Network fusion\n# Evaluation focus and outcomes","[{\"question\":\"Why is DeepADR expected to generalize to new chemical entities?\",\"answer\":\"DeepADR is designed to leverage early-stage readily available multimodal drug discovery data, combining chemical, biological, and semantic information. The reported results show robust generalization to new entities, improving reliability beyond compound-specific prior knowledge.\"}]","DeepADR - Multimodal prediction of adverse drug reaction frequency | PDF",1790605709,33]