{"id":{"repo_id":"oxford-brookes","oai_identifier":"tle:fa1965a6-1de9-47f3-adc5-06e88540af93:d6bd9758-527a-46cd-bfe2-c433766e8fca:1"},"canonical_url":"https://search.dev.ndltd.org/etd/oxford-brookes/tle:fa1965a6-1de9-47f3-adc5-06e88540af93:d6bd9758-527a-46cd-bfe2-c433766e8fca:1","repository":{"repo_id":"oxford-brookes","name":"Oxford Brookes University","base_url":"https://radar.brookes.ac.uk/radar/oai"},"display":{"title":"A novel genomic approach to multiple cancer diagnostics using mutations in blood bound EV RNA: Machine Learning Methods For Biomarker Detection and Protein Mutation Assessment","abstract":"Identification of biomarkers is critical for early detection of cancers such as pancreatic ductal adenocarcinoma (PDAC), which is often diagnosed in later stages. Extracellular vesicles (EVs) released into the blood from tumour cells make good biomarkers for this purpose by protecting their nucleotide cargo from the harsh environment, reflecting the state of the parent cell in real time. In a novel approach, blood based EV RNA mutations were analysed from patient RNAseq data for PDAC, colorectal carcinoma (CRC) and invasive lobular carcinoma (ILC). A new pipeline, with the addition of machine learning methods, was developed to calculate the mutations across the EV RNA for each patient sample across all genes, offering original panels of cancer biomarkers. Associated pathways were also highlighted based on the mutation counts across genes and samples to offer clinical insight. In an attempt to analyse the complexity of EV RNA mutations on downstream proteins, machine learning was used to develop models capable of assessing the severity of a single amino acid mutation within a protein sequence. Multiple biological parameters were built into the machine learning models in an attempt to improve on state-of-the-art methods, such as Polyphen2 . Additionally, we took an original approach to protein cross-contamination between datasets to model performance assessment. As part of this process the need for stricter validation in order to stop machine learning models extracting biases within datasets was revealed. Overall a novel pipeline and methodology for cancer biomarker discovery has been built, providing new combinations of candidate PDAC and ILC biomarkers with clinical relevance that could be expanded out to multiple cancers and clinical testing.","abstract_html":"Identification of biomarkers is critical for early detection of cancers such as pancreatic ductal adenocarcinoma (PDAC), which is often diagnosed in later stages. Extracellular vesicles (EVs) released into the blood from tumour cells make good biomarkers for this purpose by protecting their nucleotide cargo from the harsh environment, reflecting the state of the parent cell in real time. In a novel approach, blood based EV RNA mutations were analysed from patient RNAseq data for PDAC, colorectal carcinoma (CRC) and invasive lobular carcinoma (ILC). A new pipeline, with the addition of machine learning methods, was developed to calculate the mutations across the EV RNA for each patient sample across all genes, offering original panels of cancer biomarkers. Associated pathways were also highlighted based on the mutation counts across genes and samples to offer clinical insight. In an attempt to analyse the complexity of EV RNA mutations on downstream proteins, machine learning was used to develop models capable of assessing the severity of a single amino acid mutation within a protein sequence. Multiple biological parameters were built into the machine learning models in an attempt to improve on state-of-the-art methods, such as Polyphen2 . Additionally, we took an original approach to protein cross-contamination between datasets to model performance assessment. As part of this process the need for stricter validation in order to stop machine learning models extracting biases within datasets was revealed. Overall a novel pipeline and methodology for cancer biomarker discovery has been built, providing new combinations of candidate PDAC and ILC biomarkers with clinical relevance that could be expanded out to multiple cancers and clinical testing.","abstract_has_math":false,"creators":["Oliver, Jamie"],"institution":"Oxford Brookes University","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":["Pink, Ryan","Lees, Jon"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2023,"date_issued":"2023","date_published":"2023","updated_at":"2026-07-24T03:42:27Z","subjects":[],"languages":["en"],"rights":["All rights reserved"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://doi.org/10.24384/3Z7J-2417","outbound_label":"DOI","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Oliver, Jamie","Pink, Ryan","Lees, Jon"]},{"key":"dc:creator","label":"Author","values":["Oliver, Jamie"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2023"]},{"key":"dc:publisher","label":"Institution","values":["Oxford Brookes University"]},{"key":"dc:type","label":"Dc Type","values":["thesis"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["All rights reserved"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://doi.org/10.24384/3Z7J-2417","https://radar.brookes.ac.uk/radar/file/fa1965a6-1de9-47f3-adc5-06e88540af93/1/Oliver2022CancerDiagnostics.pdf"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Identification of biomarkers is critical for early detection of cancers such as pancreatic ductal adenocarcinoma (PDAC), which is often diagnosed in later stages. Extracellular vesicles (EVs) released into the blood from tumour cells make good biomarkers for this purpose by protecting their nucleotide cargo from the harsh environment, reflecting the state of the parent cell in real time. In a novel approach, blood based EV RNA mutations were analysed from patient RNAseq data for PDAC, colorectal carcinoma (CRC) and invasive lobular carcinoma (ILC). A new pipeline, with the addition of machine learning methods, was developed to calculate the mutations across the EV RNA for each patient sample across all genes, offering original panels of cancer biomarkers. Associated pathways were also highlighted based on the mutation counts across genes and samples to offer clinical insight. In an attempt to analyse the complexity of EV RNA mutations on downstream proteins, machine learning was used to develop models capable of assessing the severity of a single amino acid mutation within a protein sequence. Multiple biological parameters were built into the machine learning models in an attempt to improve on state-of-the-art methods, such as Polyphen2 . Additionally, we took an original approach to protein cross-contamination between datasets to model performance assessment. As part of this process the need for stricter validation in order to stop machine learning models extracting biases within datasets was revealed. Overall a novel pipeline and methodology for cancer biomarker discovery has been built, providing new combinations of candidate PDAC and ILC biomarkers with clinical relevance that could be expanded out to multiple cancers and clinical testing."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["A novel genomic approach to multiple cancer diagnostics using mutations in blood bound EV RNA: Machine Learning Methods For Biomarker Detection and Protein Mutation Assessment"]}]}],"canonical_facts":{"dc:contributor":["Oliver, Jamie","Pink, Ryan","Lees, Jon"],"dc:creator":["Oliver, Jamie"],"dc:date":["2023"],"dc:description":["Identification of biomarkers is critical for early detection of cancers such as pancreatic ductal adenocarcinoma (PDAC), which is often diagnosed in later stages. Extracellular vesicles (EVs) released into the blood from tumour cells make good biomarkers for this purpose by protecting their nucleotide cargo from the harsh environment, reflecting the state of the parent cell in real time. In a novel approach, blood based EV RNA mutations were analysed from patient RNAseq data for PDAC, colorectal carcinoma (CRC) and invasive lobular carcinoma (ILC). A new pipeline, with the addition of machine learning methods, was developed to calculate the mutations across the EV RNA for each patient sample across all genes, offering original panels of cancer biomarkers. Associated pathways were also highlighted based on the mutation counts across genes and samples to offer clinical insight. In an attempt to analyse the complexity of EV RNA mutations on downstream proteins, machine learning was used to develop models capable of assessing the severity of a single amino acid mutation within a protein sequence. Multiple biological parameters were built into the machine learning models in an attempt to improve on state-of-the-art methods, such as Polyphen2 . Additionally, we took an original approach to protein cross-contamination between datasets to model performance assessment. As part of this process the need for stricter validation in order to stop machine learning models extracting biases within datasets was revealed. Overall a novel pipeline and methodology for cancer biomarker discovery has been built, providing new combinations of candidate PDAC and ILC biomarkers with clinical relevance that could be expanded out to multiple cancers and clinical testing."],"dc:format":["application/pdf"],"dc:identifier":["https://doi.org/10.24384/3Z7J-2417","https://radar.brookes.ac.uk/radar/file/fa1965a6-1de9-47f3-adc5-06e88540af93/1/Oliver2022CancerDiagnostics.pdf"],"dc:language":["en"],"dc:publisher":["Oxford Brookes University"],"dc:rights":["All rights reserved"],"dc:title":["A novel genomic approach to multiple cancer diagnostics using mutations in blood bound EV RNA: Machine Learning Methods For Biomarker Detection and Protein Mutation Assessment"],"dc:type":["thesis"]},"updated_at":"2026-07-24T03:42:27Z"}