We are currently advertising the following PhD opportunity:
PhD studentship in Machine Learning / Computer Science
You can also download the Word document here:
Supervisors: Professor Wiebke Arlt & Professor Peter Tino
Application deadline: 5 March 2021
Project title: Integrated machine learning approaches to dissect the link between androgen excess and metabolic disease in women
Institution: University of Birmingham
School: Institute of Metabolism & Systems Research (IMSR) and School of Computer Science
Funding availability: Directly funded PhD project (worldwide)
Research interests/description of main research theme:
Applications are invited for a 3-year fully funded PhD Studentship starting 1 April 2021.
BACKGROUND: Polycystic ovary syndrome (PCOS) affects 10% of women worldwide. PCOS is primarily defined by irregular menstrual cycles and excess production of male hormones (androgens) and frequently associated with insulin resistance. PCOS is now emerging as a lifelong metabolic disorder with an increased risk of type 2 diabetes, hypertension, fatty liver disease and cardiovascular disease. We investigate the link between severity and pattern of androgen excess and metabolic function parameters and the risk of metabolic disease, aiming both at data-driven mechanistic discovery and identification of predictive biomarkers.
PROJECT: This multi-disciplinary PhD project will focus on the development of innovative machine learning approaches for the investigation of steroid metabolome, global non-targeted metabolome and clinical phenotype data obtained in a large cohort of women with PCOS.
ENVIRONMENT: This PhD project is embedded in a multi-disciplinary biomedical research group within the Institute of Metabolism and Systems Research (IMSR) at the University of Birmingham. We deliver a Wellcome Trust-funded translational research programme on PCOS led by Prof Wiebke Arlt (“Dissecting Androgen Excess and Metabolic Dysfunction – an integrated Systems approach to PolyCystic Ovary Syndrome”, DAISY-PCOS). This project is carried out in close collaboration with two leading experts in machine learning, Prof Peter Tino, Professor of Complex and Adaptive Systems at the University of Birmingham (Modeling of complex dynamical systems, inference in generative models), and Prof Michael Biehl, Professor of Computer Science (Clustering and classification of high-dimensional and multimodal data) at the University of Groningen, and Prof Warwick Dunn, Professor of clinical and analytical metabolomics and Director of the Phenome Centre Birmingham. All four researchers will jointly supervise this PhD project, which will be registered with the Postgraduate School of Computer Science at the University of Birmingham. We enjoy communicating our science to patients and public and encourage and provide opportunities for training and participation in public engagement activities.
TECHNIQUES IN THIS PROJECT: Unsupervised and supervised methods of machine learning and data exploration, biomedical application of machine learning approaches for mechanistic investigation, multi-steroid profiling by mass spectrometry, non-targeted global metabolome analysis, and human in vivo physiology assessments.
Person Specification
Applicants should have a strong background in a numerate discipline (Computer Science, Mathematics, Physics or a related field), with strong interest in computational modelling and machine learning, including algorithm development and applications. Good programming skills are necessary, experience with applications in life sciences and biomedical data analysis is desirable.
Applicants should be enthusiastic, self-motivated, and excellent team players. Due to the collaborative and inter-disciplinary nature of the project, outstanding communication skills and the ability and enjoyment of analytical, translational and creative thinking are required. Applicants should hold or realistically expect to obtain a Masters of Science (or equivalent) degree with excellent marks.
How to apply
Informal enquiries should be directed to Prof Peter Tino p.tino@bham.ac.uk and Professor Wiebke Arlt w.arlt@bham.ac.uk
To be considered for this studentship, please send the following documents to them as attachments to your enquiry email:
• A detailed CV, including your nationality and country of birth;
• Names and addresses of two academic referees;
• A covering letter highlighting your research experience/capabilities and your motivation to undertake a PhD within the DAISY-PCOS project;
• Copies of your degree certificates with transcripts;
• Evidence of your proficiency in the English language, if applicable.
Contact for enquiries
Professor Wiebke Arlt w.arlt@bham.ac.uk
Professor Peter Tino p.tino@bham.ac.uk
References:
Feature relevance determination for ordinal regression in the context of feature redundancies and privileged information. Pfannschmidt L, Jakob J, Hinder F, Biehl M, Tino P, Hammer B. Neurocomputing. 2020 Nov; 416: 266-279. doi: 10.1016/j.neucom.2019.12.133.
Accurate non-invasive diagnosis and staging of non-alcoholic fatty liver disease using the urinary steroid metabolome. A. Moolla, J. de Boer, D. Pavlov, A. Amin, A. Taylor, L. Gilligan, B. Hughes, J. Ryan, E. Barnes, Z. Hassan-Smith, J. Grove, G.P. Aithal, A. Verrijken, S. Francque, L. Van Gaal, M.J. Armstrong, P.N. Newsome, J.F. Cobbold, W. Arlt, M. Biehl, and J.W. Tomlinson. Alimentary Pharmacology & Therapeutics. 2020; 51(11): 1188-1197. doi: 10.1111/apt.15710.
Biomedical Applications of Prototype Based Classifiers and Relevance Learning. M. Biehl. In: Intl. Conference on Algorithms for Computational Biology AlCoB 2017. D. Figueiredo, C. Martin-Vide, D. Pratas, M.A. Vega-Rodriguez (eds.) Springer. 2017; LNCS 10252: 3-23. doi: 10.1007/978-3-319-58163-7
K. Bunte, D.J. Smith, M.J. Chappell, Z.K. Hassan-Smith, J.W. Tomlinson, W. Arlt, P. Tino: Learning Pharmacokinetic Models for in vivo Glucocorticoid Activation. Journal of Theoretical Biology, 455, pp. 222-231, DOI:10.1016/j.jtbi.2018.07.025, 2018.
Urine steroid metabolomics for the differential diagnosis of adrenal incidentalomas in the EURINE-ACT study: a prospective test validation study. Bancos I, Taylor AE, Chortis V, Sitch AJ, Jenkinson C, Davidge-Pitts CJ, Lang K, Tsagarakis S, Macech M, Riester A, Deutschbein T, Pupovac ID, Kienitz T, Prete A, Papathomas TG, Gilligan LC, Bancos C, Reimondo G, Haissaguerre M, Marina L, Grytaas MA, Sajwani A, Langton K, Ivison HE, Shackleton CHL, Erickson D, Asia M, Palimeri S, Kondracka A, Spyroglou A, Ronchi CL, Simunov B, Delivanis DA, Sutcliffe RP, Tsirou I, Bednarczuk T, Reincke M, Burger-Stritt S, Feelders RA, Canu L, Haak HR, Eisenhofer G, Dennedy MC, Ueland GA, Ivovic M, Tabarin A, Terzolo M, Quinkler M, Kastelan D, Fassnacht M, Beuschlein F, Ambroziak U, Vassiliadi DA, O’Reilly MW, Young WF Jr, Biehl M, Deeks JJ, Arlt W; ENSAT EURINE-ACT Investigators. Lancet Diabetes Endocrinol. 2020 Sep;8(9):773-781. doi: 10.1016/S2213-8587(20)30218-7. PMID: 32711725
AKR1C3-Mediated Adipose Androgen Generation Drives Lipotoxicity in Women With Polycystic Ovary Syndrome. O’Reilly MW, Kempegowda P, Walsh M, Taylor AE, Manolopoulos KN, Allwood JW, Semple RK, Hebenstreit D, Dunn WB, Tomlinson JW, Arlt W. J Clin Endocrinol Metab. 2017 Sep 1;102(9):3327-3339. doi: 10.1210/jc.2017-00947. PMID: 28645211 Molecular phenotyping of a UK population: defining the human serum metabolome. Dunn WB, Lin W, Broadhurst D, Begley P, Brown M, Zelena E, Vaughan AA, Halsall A, Harding N, Knowles JD, Francis-McIntyre S, Tseng A, Ellis DI, O’Hagan S, Aarons G, Benjamin B, Chew-Graham S, Moseley C, Potter P, Winder CL, Potts C, Thornton P, McWhirter C, Zubair M, Pan M, Burns A, Cruickshank JK, Jayson GC, Purandare N, Wu FC, Finn JD, Haselden JN, Nicholls AW, Wilson ID, Goodacre R, Kell DB.Metabolomics. 2015;11:9-26. doi: 10.1007/s11306-014-0707-1. PMID: 25598764
DAISy-PCOS 