[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-128413-en":3,"doc-seo-128413-105":31,"detail-sidebar-cat-0-en-105":92},{"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":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":28,"seo_description":14,"update_tm":29,"read_time":30},128413,8796095027276,"Valentina","https://avatar.qwps.com/avatar/d3BzX2FwX3Rlc3RfMjUxMTI2XzAxODA=",8,"Research & Report","A systems-level machine learning approach uncovers therapeutic targets in clear cell renal cell carcinoma","A systems-level machine learning framework is used to identify and refine therapeutic drug targets in clear cell renal cell carcinoma by contrasting tumour and non-tumour cellular states across multiple external datasets. The approach evaluates model discrimination with metrics including accuracy, precision, recall, F1 score and AUC using a LightGBM classifier on scRNA-seq cohorts from several regions. Supplementary analyses provide ranked gene candidates, expression directionality and functional relevance, followed by network-based refinement and drug proximity results for both FDA approved and non-approved drugs.","A systems-level machine learning approach uncovers therapeutic targets in clear cell renal cell carcinoma  \nSilas Ruhrberg Estévez1, Greta Baltusyte1,2,3,4, Gehad Youssef1,2,3*, Namshik Han1,2,3*  \n1 Milner Therapeutics Institute, University of Cambridge, UK  \n2 Cambridge Centre for AI in Medicine, University of Cambridge, UK  \n3 Cambridge Stem Cell Institute, University of Cambridge, UK  \n4 Department of Surgery, University of Cambridge, UK  \n*Corresponding authors: [gy260@cam.ac.uk](gy260@cam.ac.uk), [nh417@cam.ac.uk](nh417@cam.ac.uk)  \nSupplementary Files Information  \n• Supplementary File 1: List of all genes with AUC  \n• Supplementary File 2: Log2 fold change in expression in tumour cells compared to non-tumour cells  \n• Supplementary File 3: List of all key proteins as identified by the network analysis  \n• Supplementary File 4: Results of drug proximity for FDA approved drugs  \n• Supplementary File 5: Results of drug proximity for non-FDA approved drugs  \n\n| Dataset | Reference | Tumour | Non-tumour |\n| --- | --- | --- | --- |\n| American | Zhang et al. (2021)1 | 6903 | 8196 |\n| Chinese | Zhang et al. (2022)2 | 8545 | 46443 |\n| Lithuanian | Zvirblyte et al. (2024)3 | 3039 | 47197 |\n| Metastasis | Mei et al. (2024)4 | 1238 | 979614 |\n| Training | Li et al. (2022)5 | 6062 | 266801 |\n| TCGA Bulk | Creighton et al. (2013)6 | 533 | 73 |\n| Additional bulk | Laskar et al (2021)7 | 503 | 153 |\n\nSupplementary Table 1: Summary of datasets used in the study Tumour and non-tumour umbers correspond to number of patients for bulk RNA and number of patients for scRNA.  \n\n| Dataset | Accuracy | Precision | Recall | F1 Score | AUC |\n| --- | --- | --- | --- | --- | --- |\n| American | 0.97 | 0.99 | 0.96 | 0.97 | 0.99 |\n| Chinese | 0.97 | 0.98 | 0.98 | 0.98 | 0.99 |\n| Lithuanian | 0.98 | 0.98 | 1.00 | 0.99 | 0.99 |\n| Metastasis | 0.99 | 0.99 | 1.00 | 1.00 | 0.92 |\n\nSupplementary Table 2: Performance of tumour classification using 96 drug target genes across external datasets Summary of classification performance (accuracy, precision, recall, F1 score, and AUC) for the binary LightGBM model using the initial 96-gene target set with the tumour cells as the positive class . The classifier was evaluated on four independent scRNA-seq datasets from the United States, China, Lithuania, and a ccRCC metastatic cohort.  \n\n| Dataset | Accuracy | Precision | Recall | F1 Score | AUC |\n| --- | --- | --- | --- | --- | --- |\n| American | 0.97 | 0.95 | 0.99 | 0.97 | 0.99 |\n| Chinese | 0.97 | 0.89 | 0.93 | 0.91 | 0.99 |\n| Lithuanian | 0.98 | 0.93 | 0.73 | 0.82 | 0.99 |\n| Metastasis | 0.99 | 0.66 | 0.45 | 0.53 | 0.92 |\n\nSupplementary Table 3: Performance of tumour classification using 96 drug target genes across external datasets Summary of classification performance (accuracy, precision, recall, F1 score, and AUC) for the binary LightGBM model using the initial 96-gene target set with the non-tumour cells as the positive class . The classifier was evaluated on four independent scRNA-seq datasets from the United States, China, Lithuania, and a ccRCC metastatic cohort.  \n\n| Gene\u003Cbr>name | Expression | Function in cancer |\n| --- | --- | --- |\n| ENO1 | upregulated | Glycolytic enzyme, has been implicated in metabolic reprogramming, immune evasion and apoptosis inhibition8 |\n| CD52 | downregulated | CD52 is a glycoprotein found on the surface of immune cells that has been shown to regulate immune invasion9 |\n| PTPRC | downregulated | Inhibits JAK family kinases10 |\n| SPP1 | upregulated | Induction of epithelial-mesenchymal transition and drug resistance11 |\n| DAB2 | upregulated | Loss is associated with MAPK, Wnt and TGFβ signalling that drives tumour progression12 |\n| HINT1 | upregulated | Potentially limits CD4+ T cell infiltration13 |\n| HLA-E | downregulated | Negative immunomodulation by overexpression14 |\n| TIMP1 | upregulated | Promotes tumorigenesis via epithelialmesenchymal transition in ccRCC15 |\n| PGK1 | upregulated | Promotes tumorigenesis and sorafeni","cbCaifRIaA5Zrs8N","https://ap.wps.com/l/cbCaifRIaA5Zrs8N","pdf",1426339,3,1,10,"English","en",105,"# Supplementary Files Information\n## Gene lists and AUC results\n## Expression fold-change comparisons\n## Network key proteins\n## Drug proximity analyses (FDA approved and non-approved)\n# Dataset Resources and Cohort Summaries\n## Tumour vs non-tumour patient counts across studies\n# Model Performance for Tumour Classification\n## Accuracy, precision, recall, F1 score, and AUC\n# Model Performance for Non-tumour Classification\n## Accuracy, precision, recall, F1 score, and AUC\n# Refined Drug Target Genes and Functional Annotations\n## Expression direction and cancer-related functions\n# Functional Characterization of Refined Drug Targets\n## Refined target set performance across external datasets\n# Classification Performance for Refined 17-Gene Target Set\n## External dataset metric summary","[{\"question\":\"What is the main aim of the systems-level machine learning workflow in this study?\",\"answer\":\"To discover and refine therapeutic targets for clear cell renal cell carcinoma by learning patterns that distinguish tumour from non-tumour states and by integrating network-based evidence.\"},{\"question\":\"How is classification performance evaluated for tumour vs non-tumour groups?\",\"answer\":\"Using LightGBM models assessed on accuracy, precision, recall, F1 score and AUC across multiple independent scRNA-seq datasets.\"},{\"question\":\"What supplementary outputs are provided to support target discovery?\",\"answer\":\"Supplementary files include gene lists with AUC, log2 expression fold changes, key proteins from network analysis, and drug proximity results for both FDA approved and non-approved drugs.\"}]","A systems-level machine learning approach uncovers therapeutic targets in clear cell renal cell carcinoma | PDF",1785947375,25,{"code":4,"msg":32,"data":33},"ok",{"site_id":25,"language":24,"slug":34,"title":13,"keywords":35,"description":14,"schema_data":36,"social_meta":87,"head_meta":89,"extra_data":91,"updated_unix":29},"a-systems-level-machine-learning-approach-uncovers-therapeutic-targets-in-clear-cell-renal-cell-carcinoma","",{"@graph":37,"@context":86},[38,54,69],{"@type":39,"itemListElement":40},"BreadcrumbList",[41,45,49,51],{"item":42,"name":43,"@type":44,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":46,"name":47,"@type":44,"position":48},"https://docshare.wps.com/document/","Document",2,{"item":50,"name":12,"@type":44,"position":20},"https://docshare.wps.com/document/research-report/",{"item":52,"name":13,"@type":44,"position":53},"https://docshare.wps.com/document/a-systems-level-machine-learning-approach-uncovers-therapeutic-targets-in-clear-cell-renal-cell-carcinoma/128413/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":24,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":42,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-27","2026-08-05",true,{"@type":66,"interactionType":67,"userInteractionCount":20},"InteractionCounter",{"@type":68},"ViewAction",{"@type":70,"mainEntity":71},"FAQPage",[72,78,82],{"name":73,"@type":74,"acceptedAnswer":75},"What is the main aim of the systems-level machine learning workflow in this study?","Question",{"text":76,"@type":77},"To discover and refine therapeutic targets for clear cell renal cell carcinoma by learning patterns that distinguish tumour from non-tumour states and by integrating network-based evidence.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How is classification performance evaluated for tumour vs non-tumour groups?",{"text":81,"@type":77},"Using LightGBM models assessed on accuracy, precision, recall, F1 score and AUC across multiple independent scRNA-seq datasets.",{"name":83,"@type":74,"acceptedAnswer":84},"What supplementary outputs are provided to support target discovery?",{"text":85,"@type":77},"Supplementary files include gene lists with AUC, log2 expression fold changes, key proteins from network analysis, and drug proximity results for both FDA approved and non-approved drugs.","https://schema.org",{"og:url":52,"og:type":88,"og:title":13,"og:site_name":59,"og:description":14},"article",{"robots":90,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,135],{"id":21,"doc_module":4,"doc_module_name":47,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":48,"doc_module":4,"doc_module_name":47,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":47,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":47,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":47,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":47,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":47,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":47,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":47,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":22,"doc_module":4,"doc_module_name":47,"category_name":133,"show_sort_weight":22,"slug":134},"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":47,"category_name":137,"show_sort_weight":107,"slug":138},19,"General","general"]