[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-335539-105":59,"doc-detail-335539-en":131},{"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":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","deside-a-unified-deep-learning-approach-for-cellular-deconvolution-of-tumor-microenvironment","DeSide - A unified deep learning approach for cellular deconvolution of tumor microenvironment","","Cellular deconvolution using bulk RNA sequencing (RNA-seq) offers a cost-effective alternative to flow cytometry and single-cell RNA-seq, but tumor microenvironments remain difficult to analyze due to extensive heterogeneity among tumors. DeSide presents a unified deep learning framework that estimates proportions of 16 cell types and subtypes in solid tumors. It integrates biological pathways and prioritizes noncancerous cell types to bypass highly variable gene expression in cancer cells. Using scRNA-seq references from six cancer types and 185 cancer cell lines across 22 cancers, DeSide builds a training set aligned with TCGA bulk RNA-seq, outperforming existing methods for tumor purity and noncancerous cell fractions across 19 cancer types, supported by external datasets. The model further enables analysis of combinatorial cell-type pairs for patient stratification.",{"@graph":69,"@context":123},[70,84,106],{"@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/deside-a-unified-deep-learning-approach-for-cellular-deconvolution-of-tumor-microenvironment/335539/",{"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/deside-a-unified-deep-learning-approach-for-cellular-deconvolution-of-tumor-microenvironment/335539.png","ImageObject",300,407,{"name":92,"@type":93},"Quinn","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-11","2026-09-21",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",15,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What problem does DeSide address in tumor microenvironment analysis?","Question",{"text":113,"@type":114},"DeSide addresses difficulty in inferring cellular composition from bulk RNA-seq because tumors show strong heterogeneity within and across tumors and because cancer cells can have highly variable gene expression profiles.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does DeSide improve on existing cellular deconvolution methods?",{"text":118,"@type":114},"DeSide uses a unified deep learning model that integrates biological pathways, emphasizes noncancerous cell types first, and constructs a high-quality training set using scRNA-seq references guided toward TCGA bulk RNA-seq-like expression patterns.",{"name":120,"@type":111,"acceptedAnswer":121},"What outputs and downstream uses does DeSide enable?",{"text":122,"@type":114},"DeSide estimates tumor purity and proportions of noncancerous cell types across multiple cancer types, and it supports identification and analysis of combinatorial cell-type pairs to stratify patients into prognostically meaningful groups.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},335539,1790160001,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":145,"read_time":41,"preview_page_count":4},962075114765,"https://ap-avatar.wpscdn.com/davatar_a8503ba1806abce46bf441b54a3ca4cd","RESEARCH ARTICLE  \nBIOPHYSICS AND COMPUTATIONAL BIOLOGY  \n OPEN ACCESS  \nDeSide: A unified deep learning approach for cellular deconvolution of tumor microenvironment  \nXin Xionga,1, Yerong Liub,1, Dandan Pub, Zhu Yangc , Zedong Bid, Liang Tiana,c,e,f,2, and Xuefei Lib,2 Affiliations are included on p. 11.  \nEdited by Jens Nielsen, BioInnovation Institute, Hellerup, Denmark; received April 9, 2024; accepted September 23, 2024  \nCellular deconvolution via bulk RNA sequencing (RNA-seq) presents a cost-effective and efficient alternative to experimental methods such as flow cytometry and single-cell RNA-seq (scRNA-seq) for analyzing the complex cellular composition oftumor microenvironments. Despite challenges due to heterogeneity within and among tumors, our innovative deep learning–based approach, DeSide, shows exceptional accuracy in estimating the proportions of 16 distinct cell types and subtypes within solid tumors. DeSide integrates biological pathways and assesses noncancerous cell types first, effectively sidestepping the issue of highly variable gene expression profiles (GEPs) associated with cancer cells. By leveraging scRNA-seq data from six cancer types and 185 cancer cell lines across 22 cancer types as references, our method introduces distinctive sampling and filtering techniques to generate a high-quality training set that closely replicates real tumor GEPs, based on The Cancer Genome Atlas (TCGA) bulk RNA-seq data. With this model and high-quality training set, DeSide outperforms existing methods in estimating tumor purity and the proportions of noncancerous cells within solid tumors. Our model precisely predicts cellular compositions across 19 cancer types from TCGA and proves its effectiveness with multiple additional external datasets. Crucially, DeSide enables the identification and analysis of combinatorial cell type pairs, facilitating the stratification of cancer patients into prognostically significant groups. This approach not only provides deeper insights into the dynamics of tumor biology but also highlights potential therapeutic targets by underscoring the importance of specific cell type or subtype interactions.  \ncellular deconvolution | single-cell RNA sequencing | deep learning | bulk RNA sequencing | tumor microenvironment  \nSolid tumors are complex mixtures of various cell types, including cancerous cells and a diverse range of cells in the tumor microenvironment (TME), such as cancer-associated fibroblasts (CAFs), endothelial cells, tumor-associated macrophages, and tumor-infiltrating lymphocytes (TILs) (1–5). The presence and quantities of these cells, particularly TILs, have been associated with patient outcomes across multiple cancers (6, 7), underscoring the significance of precisely evaluating tumor cellular composition for clinical understanding therapeutic interventions (3, 6–9) .  \nCommonly, the cellular composition of tumors has been investigated through methods like flow cytometry (10), single-cell RNA sequencing (scRNA-seq) (11), or immunohistochemistry (IHC) analysis of the tumor section images (12) . Nevertheless, these approaches have limitations related to cost, coverage of cellular types, or extraction efficiency. In contrast, bulk RNA-seq presents a cost-effective alternative, leveraging publicly available data from projects like The Cancer Genome Atlas (TCGA) (13), to analyze abroad spectrum of tumor samples.  \nThe considerable variability in gene expression profiles (GEPs) among cancerous cells, amplified by genetic and epigenetic disruptions (14–16), poses a challenge in identifying universal markers for cancer cells. This stronger variability (SI Appendix, Fig. S1), along with limited scRNA-seq data for some cancer types, makes predicting the proportions of cancerous cells directly much more challenging than predicting noncancerous cells.  \nSeveral computational methods, including EPIC (17), MuSiC (18), and CIBERSORT/ CIBERSORTx (19, 20), have been developed to infer the c","cbCaittnIWyuQiPf","https://ap.wps.com/l/cbCaittnIWyuQiPf","pdf",5736880,12,"English","# Significance\n## Cellular deconvolution challenges and DeSide’s goal\n## Key methodology and data strategy\n## Performance evaluation and clinical relevance","[{\"question\":\"What problem does DeSide address in tumor microenvironment analysis?\",\"answer\":\"DeSide addresses difficulty in inferring cellular composition from bulk RNA-seq because tumors show strong heterogeneity within and across tumors and because cancer cells can have highly variable gene expression profiles.\"},{\"question\":\"How does DeSide improve on existing cellular deconvolution methods?\",\"answer\":\"DeSide uses a unified deep learning model that integrates biological pathways, emphasizes noncancerous cell types first, and constructs a high-quality training set using scRNA-seq references guided toward TCGA bulk RNA-seq-like expression patterns.\"},{\"question\":\"What outputs and downstream uses does DeSide enable?\",\"answer\":\"DeSide estimates tumor purity and proportions of noncancerous cell types across multiple cancer types, and it supports identification and analysis of combinatorial cell-type pairs to stratify patients into prognostically meaningful groups.\"}]","DeSide - A unified deep learning approach for cellular deconvolution of tumor microenvironment | PDF",1790011363]