(Bicentenary) Developing machine learning-based methods for multi-omic evaluation of breast cancer metastasis at The University of Manchester

IntroductionBreast cancers are the most common cancers in women. Estrogen Receptor (ER) plays a major role in breast cancer growth. ER-targeting therapies were clinically successful to stop the cancer from growing further. However, around 20% patients develop resistance to existing therapies or cancer spreads to other organs (metastasis), which leads to poor survival.Interestingly, our previous study along with another study(1, 2) identified the loss of a chromatin remodelling protein ARID1A to be a critical factor which can promote resistance to ER-targeting therapies. ARID1A mutations are enriched in 12% of metastatic breast cancer patients compared to primary tumours(3). Importantly, ARID1A loss leads to increased migration and metastasis of ER+ breast cancer cells to other organs. Aims and approachWe hypothesise that ARID1A loss leads to aggressive and metastasized tumours by interacting with tumour microenvironment in primary tumours and cells in the host organs when disseminated. ARID1A loss would also lead to reprogrammed activation of distinct transcription factors and coactivators, modulated by stromal cells/host organs. To address this hypothesis, we need to identify if the patients have mutations in ARID1A and interrogate tumour-stromal and cancer-host interactions using integrated multi-omic approaches. Hence this project will allow the student to work with a multidisciplinary team to be trained in single cell transcriptomics, epigenomics and proteomics, and integration of datasets using computational approaches. The project will utilise primary and metastatic patient samples from tissue biobanks, but these have limited number of cells especially the metastatic samples are originated from biopsies. Hence, the project will examine if limited approaches can be utilised through multi-omic integration. Using this workflow, the student will identify the detailed mechanisms of ARID1A in regulating metastasis, by performing single nuclear transcriptomics, epigenomics and proteomics and deriving the genetic mutations, cell-cell interactions and epigenetic activation of transcription factors using bespoke computational approaches. As this project will be largely bioinformatics-driven, graduates with bioinformatics background are strongly encouraged to apply. Impact:This robust interdisciplinary project aims to hone the student in a diverse set of skills integrating breast cancer epigenetics with single nuclear next generation sequencing and machine learning and make significant strides in addressing critical clinical knowledge gaps. The student will be trained to achieve better understanding of the gene regulation during cancer spread (disease mechanisms), further leading to predictive gene expression patterns of metastasis and developing therapeutic interventions against the identified epigenetic regulators, aiding in patient stratification and better health care. Entry RequirementsApplicants are expected to hold (or about to obtain) a minimum upper second-class undergraduate honours degree (or equivalent) in any of the following fields such as computer science, data science, bioinformatics, machine/deep learning, Artificial intelligence.  Research experience in Python, R and machine learning approaches is desirable. Application GuidanceCandidates must contact the primary supervisor before applying to discuss their interest in the project and assess their suitability.  Apply directly via this link: https://tinyurl.com/ycweuusx or on the online application portal, select FBMH Bicentenary PhD as the programme of study.You may apply for up to two projects within this scheme. To do so, submit a single online application listing both project titles and the names of both main supervisors in the relevant sections.  Please ensure that your application includes all required supporting documents:Curriculum Vitae (CV)  Supporting Statement  Academic Certificates and TranscriptsIncomplete or late applications will not be considered. Further details are available on our website: Bicentenary PhD Studentships and Fellowships | Biology, Medicine and Health | University of Manches…