[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-119749-en":3,"doc-seo-119749-105":30,"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":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},119749,549758146520,"Patrick","https://ap-avatar.wpscdn.com/avatar/80002397d8c0411e94?_k=1775819394049821470",8,"Research & Report","LARGE-SCALE AND PAN-CANCER MULTI-OMIC ANALYSES WITH MACHINE LEARNING - Thesis Abstract","Multi-omic data analysis underpins molecular biology research, and distinct omic modalities can reveal patterns not observable in any single layer. Leveraging mass-spectrometry enabled proteomics, this thesis develops machine learning methods for integrative analysis of multi-omic cancer cell line and tissue datasets. It benchmarks existing integration approaches for drug-response prediction and cancer classification, constructs a pan-cancer proteomic map across multiple cancer types, and introduces DeeProM and DeePathNet for improved predictive modeling with interpretable biomarker discovery.","LARGE-SCALE AND PAN-CANCER MULTI-OMIC ANALYSES WITH MACHINE  \nLEARNING  \nZhaoxiang Cai  \nA thesis submitted in fulfilment of the requirements for the degree of  \nDoctor of Philosophy  \nChildren’s Medical Research Institute Faculty of Medicine and Health The University of Sydney  \nJan 2023  \nKeywords  \nCancer; Multi-omic data integration; Drug response, CRISPR-Cas9; Cancer vulnerabilities; Machine learning; Deep learning.  \nThesis Abstract  \nMulti-omic data analysis has been foundational in many fields of molecular biology, including cancer research. Investigation of the relationship between different omic data types reveals patterns that cannot otherwise be found in a single data type alone. With recent technological advancements in mass spectrometry (MS), MS-based proteomics has enabled the quantification of thousands of proteins in hundreds of cell lines and human tissue samples. The proteome of these lines and samples have provided additional insights into disease biology beyond the genome and transcriptome. This thesis presents several machine learning-based methods that facilitate the integrative analysis of multi-omic data.  \nFirst, we reviewed five existing multi-omic data integration methods and performed a benchmarking analysis, using a large-scale multi-omic cancer cell linedataset. We evaluated the performance of these machine learning methods for drug response prediction and cancer type classification. Our result provides recommendations to researchers regarding optimal machine learning method selection for their applications.  \nSecond, we generated a pan-cancer proteomic map of 949 cancer cell lines across 40 cancer types and developed a machine learning method DeeProM to analyse the multi-omic information of these lines. DeeProM identifies 8,498 proteins with evidence of cell types, protein-protein interaction, and broad post-transcriptional regulation. It also discovers protein-specific biomarkers of drug response and gene essentiality. The predictive performance of this dataset using machine learning was comparable with the RNA-seq and an external proteomic dataset. Further analysis demonstrated that a random subset of 1,500 proteins had a limited impact on predictive  \nperformance, consistent with protein networks being highly connected and coregulated. This pan-cancer proteomic map (ProCan-DepMapSanger) is now publicly available and represents a major resource for the scientific community, for biomarker discovery and for the study of fundamental aspects of protein regulation.  \nThird, we focused on publicly available multi-omic datasets of both cancer cell lines and human tissue samples and developed a Transformer-based deep learning method, DeePathNet, which integrates human knowledge with machine intelligence. DeePathNet incorporates cancer pathway knowledge into its network design by grouping omic data. A Transformer encoder was utilised to dynamically model the interdependency between cancer pathways, further improving the predictive performance. We applied DeePathNet on three evaluation tasks, namely drug response prediction, cancer type classification and breast cancer subtype classification. For a wide range of experiments, DeePathNet achieved better predictive performance than other methods that do not incorporate knowledge of cancer pathways. We used SHapley Additive exPlanations (SHAP) and Layer-wise Relevance Propagation (LRP) for model explanation and identify several key omic features and pathways that were related to breast cancer subtype classification.  \nTaken together, our analyses and methods allowed more accurate cancer diagnosis and prognosis.  \nStatement of Originality  \nThis is to certify that to the best of my knowledge, the content of this thesis is my own work. This thesis has not been submitted for any degree or other purposes.  \nI certify that the intellectual content of this thesis is the product of my own work and that all the assistance received in preparing this thesis and sourc","cbCainQStiJwxhfi","https://ap.wps.com/l/cbCainQStiJwxhfi","pdf",5637707,1,169,"English","en",105,"# Thesis Abstract\n## Review and benchmarking of multi-omic integration methods\n## Pan-cancer proteomic mapping and DeeProM\n## Transformer-based DeePathNet with pathway knowledge\n## Model explainability and biomarker discovery\n# Statement of Originality\n# Acknowledgements\n# Authorship Attribution Statement","[{\"question\":\"What does the thesis focus on regarding multi-omic cancer data?\",\"answer\":\"It targets integrative analysis across multiple omic data types, using machine learning to uncover patterns and improve tasks such as drug-response prediction and cancer classification.\"},{\"question\":\"How do DeeProM and the pan-cancer proteomic map contribute to the research?\",\"answer\":\"DeeProM is developed to analyze multi-omic information from 949 cancer cell lines, and the resulting pan-cancer proteomic map identifies thousands of proteins and supports biomarker discovery for drug response and gene essentiality.\"},{\"question\":\"How does DeePathNet differ from other approaches in predictive performance and interpretability?\",\"answer\":\"DeePathNet uses a Transformer architecture that incorporates cancer pathway knowledge into network design, achieving stronger performance on evaluation tasks and using SHAP and LRP to explain key features and pathways relevant to breast cancer subtypes.\"}]","LARGE-SCALE AND PAN-CANCER MULTI-OMIC ANALYSES WITH MACHINE LEARNING - Thesis Abstract | PDF",1785726103,426,{"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":87,"head_meta":89,"extra_data":91,"updated_unix":28},"large-scale-and-pan-cancer-multi-omic-analyses-with-machine-learning-thesis-abstract","",{"@graph":36,"@context":86},[37,54,69],{"@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/large-scale-and-pan-cancer-multi-omic-analyses-with-machine-learning-thesis-abstract/119749/",4,{"url":52,"name":13,"@type":55,"author":56,"headline":13,"publisher":58,"fileFormat":61,"inLanguage":23,"description":14,"dateModified":62,"datePublished":63,"encodingFormat":61,"isAccessibleForFree":64,"interactionStatistic":65},"DigitalDocument",{"name":9,"@type":57},"Person",{"url":41,"name":59,"@type":60},"DocShare","Organization","application/pdf","2026-08-04","2026-08-03",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 does the thesis focus on regarding multi-omic cancer data?","Question",{"text":76,"@type":77},"It targets integrative analysis across multiple omic data types, using machine learning to uncover patterns and improve tasks such as drug-response prediction and cancer classification.","Answer",{"name":79,"@type":74,"acceptedAnswer":80},"How do DeeProM and the pan-cancer proteomic map contribute to the research?",{"text":81,"@type":77},"DeeProM is developed to analyze multi-omic information from 949 cancer cell lines, and the resulting pan-cancer proteomic map identifies thousands of proteins and supports biomarker discovery for drug response and gene essentiality.",{"name":83,"@type":74,"acceptedAnswer":84},"How does DeePathNet differ from other approaches in predictive performance and interpretability?",{"text":85,"@type":77},"DeePathNet uses a Transformer architecture that incorporates cancer pathway knowledge into network design, achieving stronger performance on evaluation tasks and using SHAP and LRP to explain key features and pathways relevant to breast cancer subtypes.","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":24},{"code":4,"msg":5,"data":93},[94,98,102,106,111,116,121,124,129,132,136],{"id":20,"doc_module":4,"doc_module_name":46,"category_name":95,"show_sort_weight":96,"slug":97},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},"Literature",80,"literature",{"id":53,"doc_module":4,"doc_module_name":46,"category_name":103,"show_sort_weight":104,"slug":105},"Exam",70,"exam",{"id":107,"doc_module":4,"doc_module_name":46,"category_name":108,"show_sort_weight":109,"slug":110},5,"Comic",60,"comic",{"id":112,"doc_module":4,"doc_module_name":46,"category_name":113,"show_sort_weight":114,"slug":115},6,"Technology",50,"technology",{"id":117,"doc_module":4,"doc_module_name":46,"category_name":118,"show_sort_weight":119,"slug":120},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":122,"slug":123},30,"research-report",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":126,"show_sort_weight":127,"slug":128},9,"Religion & Spirituality",20,"religion-spirituality",{"id":127,"doc_module":4,"doc_module_name":46,"category_name":130,"show_sort_weight":127,"slug":131},"World Cup","world-cup",{"id":133,"doc_module":4,"doc_module_name":46,"category_name":134,"show_sort_weight":133,"slug":135},10,"Lifestyle","lifestyle",{"id":137,"doc_module":4,"doc_module_name":46,"category_name":138,"show_sort_weight":107,"slug":139},19,"General","general"]