Sylvia Richardson holds the Chair of Biostatistics in the Department of Epidemiology and Public Health at Imperial since 2000 and heads the Biostatistics group, one of the largest groups of academics, postdoctoral students and PhD students working on biostatistics and statistical genetics in the UK. She was previously "Directeur de Recherches" at the French Research Institute for Medical Research INSERM, where she held research positions for 20 years.
I have a Doctorat Es Sciences from the University of Paris XI, a PhD in Probability Theory from the University of Nottingham, and have held lectureship positions at Warwick University and the University of Paris V. During the 80s, I worked on the theory of spatial processes and was involved in developing methods and applications in geographical epidemiology. In the 90s, I became fully involved in the development of generic Bayesian methodology and explaining its basis and its impact to a wider audience, co-editing the best selling book "MCMC in Practice" (1996). I took an active part in the European Science Foundation (ESF) network and subsequent programme on"Highly Structured Stochastic Systems (HSSS) (1997-2000) for which I became Vice-chairman. I have worked extensively on advancing Bayesian methodology and its application in the biomedical field. In 2000, I moved to the UK to take up the Chair of Biostatistics at Imperial College. While continuing to develop Bayesian methodology in spatial epidemiology and observational studies, I started a new line of research stimulated by the statistical challenges of new biotechnologies. Building on my experience of Bayesian modelling and computation, I obtained funding from the EPSRC-BBSRC "Exploiting Genomics Initiative" to start a comprehensive research programme on Bayesian statistical genomics.
At Imperial, I am co-director of an ESRC Research Node BIAS (Bayesian methods for combining multiple Individual and Aggregate data Sources in observational studies), PI and member of Executive Committee of the recently created MRC-EPA Centre for Environment and Health, and I am involved in several collaborative projects in genomics. I have a broad range of external activities, notably on scientific boards with AFSSET (Paris), INSERM Research Assessment, the Health Effects Institute (Boston), and the Centre of "Statistics for Innovation" (Oslo).
In 2009, I was awarded the Guy's medal in Silver from the Royal Statistical Society.
The research in my group uses innovative modelling and computational strategies, principally in the Bayesian paradigm, where the complexity of the data sets analysed, ranging from intricate patterns of dependence, mis-measurement, multidimensional and multivariate aspects, are fully faced and exploited at key steps of the analysis strategy. The methods we used cover several different areas, which include:
- Mixture and clustering models, and their use in spatial analysis, epidemiology and genomics
- Sparse multivariate modelling and variable selection problems for very large sets of variables, applications in genetics
- Bayesian Integrative Genomics
- Incorporating uncertainty and measurement error in risk estimation of health effects
- Bayesian synthesis for combining multiple data sources in observational epidemiology
- Stochastic algorithms and Approximate Bayesian computations.
Our group also runs short courses on Bayesian hierarchical modelling and WinBUGS.
et al., 2016, Phenotype Similarity Regression for Identifying the Genetic Determinants of Rare Diseases., Am J Hum Genet, Vol:98, Pages:490-499
et al., 2016, MT-HESS: an efficient Bayesian approach for simultaneous association detection in OMICS datasets, with application to eQTL mapping in multiple tissues, Bioinformatics, Vol:32, ISSN:1367-4803, Pages:523-532
et al., 2016, Multidimensional analysis of the effect of occupational exposure to organic solvents on lung cancer risk: the ICARE study, Occupational and Environmental Medicine, Vol:73, ISSN:1351-0711, Pages:368-377
Papathomas M, Richardson S, 2016, Exploring dependence between categorical variables: Benefits and limitations of using variable selection within Bayesian clustering in relation to log-linear modelling with interaction terms., J Stat Plan Inference, Vol:173, ISSN:0378-3758, Pages:47-63
et al., 2015, Bayesian regression discontinuity designs: incorporating clinical knowledge in the causal analysis of primary care data., Stat Med, Vol:34, Pages:2334-2352