[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-327716-105":59,"doc-detail-327716-en":130},{"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":123,"head_meta":125,"extra_data":127,"updated_unix":129},105,"en","enhancing-drug-and-cell-line-representations-via-contrastive-learning-for-improved-anti-cancer-drug-prioritization","Enhancing drug and cell line representations via contrastive learning for improved anti-cancer drug prioritization","","Precision oncology uses omics sequence analysis but faces two bottlenecks: rapidly selecting the best regimen from large, per-patient datasets and learning effective drug–cell line patterns when drug–response data are limited or incomplete. This article proposes contrastive learning to strengthen learned drug and cell line representations by preserving relationship structures tied to drug mechanisms of action and cell line cancer types. Compared with a state-of-the-art baseline, the method improves predictive performance and yields classifiers that rely more evenly on drug- and cell-line-derived features, enabling more balanced, resistance-informed drug prioritization.",{"@graph":69,"@context":122},[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":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/enhancing-drug-and-cell-line-representations-via-contrastive-learning-for-improved-anti-cancer-drug-prioritization/327716/",{"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/enhancing-drug-and-cell-line-representations-via-contrastive-learning-for-improved-anti-cancer-drug-prioritization/327716.png","ImageObject",300,407,{"name":92,"@type":93},"Eliana","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-10-23","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":24},"InteractionCounter",{"@type":104},"ViewAction",{"@type":106,"mainEntity":107},"FAQPage",[108,114,118],{"name":109,"@type":110,"acceptedAnswer":111},"Why is drug prioritization in precision oncology difficult for clinicians and models?","Question",{"text":112,"@type":113},"Each patient generates large omics datasets, making it hard to quickly identify the best regimen. Limited data availability also restricts computational methods from learning reliable patterns for effective drug–cell line pairs.","Answer",{"name":115,"@type":110,"acceptedAnswer":116},"What does the proposed method aim to improve in drug and cell line representations?",{"text":117,"@type":113},"It uses contrastive learning to enhance learned representations by preserving relationship structures linked to drug mechanisms of action and the cancer types of cell lines.",{"name":119,"@type":110,"acceptedAnswer":120},"How do the learned representations affect prediction behavior compared with a state-of-the-art method?",{"text":121,"@type":113},"Classifiers using the learned representations show more balanced reliance on drug-derived and cell-line-derived features, supporting more personalized drug prioritization informed by signals related to drug resistance.","https://schema.org",{"og:url":83,"og:type":124,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":126,"canonical":83},"index,follow",{"doc_id":128,"site_id":62},327716,1790639256,{"code":4,"msg":5,"data":131},{"doc_id":128,"user_id":132,"nickname":92,"user_avatar":133,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":134,"file_id":135,"file_url":136,"file_type":137,"file_size":138,"view_count":24,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":139,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145,"preview_page_count":4},4398048949847,"https://ap-avatar.wpscdn.com/avatar/400002536579ef2da7f?_k=1778318612642679267","npj | precision oncology Article  \n\n| Published in partnership with The Hormel Institute, University of Minnesota |  | |\n| --- | --- | --- |\n| [https://doi.org/10.1038/s41698-024-00589-8](https://doi.org/10.1038/s41698-024-00589-8) |  |  |\n| Enhancing drug and cell line representations via contrastive learning for improved anti-cancer drug prioritization\u003Cbr> Check for updates |  |  |\n| Patrick J. Lawrence 1, Benjamin Burns2 & Xia Ning 1,2,3  |  |  |\n| Due to cancer’s complex nature and variable response to therapy, precision oncology informed by omics sequence analysis has become the current standard of care. However, the amount of data produced for each patient makes it difficult to quickly identify the best treatment regimen. Moreover, limited data availability has hindered computational methods’ abilities to learn patterns associated with effective drug-cell line pairs. In this work, we propose the use of contrastive learning to improve learned drug and cell line representations by preserving relationship structures associated with drug mechanisms of action and cell line cancer types. In addition to achieving enhanced performance relative toa state-of-the-art method, wefind that classifiers using our learned representations exhibit amore balanced reliance on drug-and cell line-derived features when making predictions. This facilitates more personalized drug prioritizations that are informed by signals related to drug resistance. |  |  |\n| Cancer, the leading cause of death worldwide, remains a challenge to treat due to its complex nature and variable response to therapy, even among patients with the same cancer type. Omics, which describes the collective and comprehensive analysis of biomolecular data, is being leveraged to conduct precision oncology. Transcriptomics, a subfield of omics that studies gene expression by quantifying relative levels of RNA molecules, has been an especially valuable tool. Oncologists use transcriptomics—comparing normal and tumor cells—to identify changes in gene expression. These alterations are used to pinpoint the molecular process(es) driving tumorigenesis in each patient, allowing clinicians to develop personalized treatment recommendations. However, because RNA sequencing (RNAseq) measures the expression of more than 20,000 genes, manually evaluating each patient’s data to determine the best treatment is neither scalable nor pragmatic.\u003Cbr>Instead, machine learning models have been applied to omics data to predict cellular sensitivities to drug candidates. RefDNN1, one state-of-theart(SOTA)method, represents drugs via their structural similarity toa set of reference drugs. It then leverages these reference drugs to produce cell line | representations: each dimension is the output of an Elastic Net2 model trained on transcriptomic data to predict a cell line’s sensitivity to a distinct reference drug. During evaluation, RefDNN applies a dense neural network (DNN) to the Hadamard product of drug and cell line representations to predict cancer drug response (CDR). However, RefDNN is limited by data quality: it requires complete CDR data for all reference drug-cell line pairs during training. Other recent SOTA methods, such as SubCDR3, have explored the use of expert annotations genes, explicitly denoting the role each of the genes in their subset played in the tumorigenesis of each cancer type in their data set. This method also relies on the use of CDR values as‘side information’. While it improves on RefDNN, in that it can be trained on incomplete CDR data, SubCDR is only able to accurately predict CDR response for new combinations of cell lines and drugs seen during training. Furthermore, manually labeling genes by their role is both time-consuming and may introduce bias into the model.\u003Cbr>DeepDSC4, another SOTA method, also leverages transcriptomic data to produce cell line representations. It does so via the latent embeddingsofa pretrained autoencoder(AE). By encoding cell lines’ transcripto","cbCaibaOYKi9Ck7c","https://ap.wps.com/l/cbCaibaOYKi9Ck7c","pdf",2046779,14,"English","# Introduction\n## Challenges in precision oncology and omics-based drug response prediction\n## Limitations of existing SOTA methods\n# Proposed approach\n## Siamese neural network with contrastive learning (SiamCDR)\n## Contrastive loss to improve embeddings\n# Model architectures and evaluation context\n## Drug and cell line representation components","[{\"question\":\"Why is drug prioritization in precision oncology difficult for clinicians and models?\",\"answer\":\"Each patient generates large omics datasets, making it hard to quickly identify the best regimen. Limited data availability also restricts computational methods from learning reliable patterns for effective drug–cell line pairs.\"},{\"question\":\"What does the proposed method aim to improve in drug and cell line representations?\",\"answer\":\"It uses contrastive learning to enhance learned representations by preserving relationship structures linked to drug mechanisms of action and the cancer types of cell lines.\"},{\"question\":\"How do the learned representations affect prediction behavior compared with a state-of-the-art method?\",\"answer\":\"Classifiers using the learned representations show more balanced reliance on drug-derived and cell-line-derived features, supporting more personalized drug prioritization informed by signals related to drug resistance.\"}]","Enhancing drug and cell line representations via contrastive learning for improved anti-cancer drug prioritization | PDF",1789977897,35]