Professor Jean-Baptiste Cazier

Jean-Baptiste Cazier


Turing Fellow

Partner Institution


Prof. Jean-Baptiste Cazier, chair of Bioinformatics, Director of the Centre for Computational Biology, UoB, is originally trained in mathematical modelling. He joined the field of human genetics when introduced to this fast-evolving field in an academic spirit of research and excellence in Iceland at deCode Genetics, developing further methods combining linkage and case-control association to identify genes responsible for common complex diseases. While employed by Cancer Research UK, he worked alongside experts in most aspects of bioinformatics and biostatistics in the context of cancer and collaborated with scientists and clinicians on genome-wide association, copy number variations or high-throughput sequencing, especially of colorectal cancer and leukaemia which led to the identification of many genomic susceptibility variants conferring higher, susceptibility and progression, risk of various cancers.

Joining the Wellcome Trust Centre for Human Genetics in Oxford, where he developed new methods such as a population genetics analysis for genome-wide association in a Middle Eastern cohort. These further studies in population stratification and admixture mapping to perform more accurate analysis across heterogeneous cohorts led to new analytical methods being developed in collaboration with the Department of Statistics. After acting as joint Head of the Bioinformatics and Statistical Genetics Core, he supervised the development of analytical approaches and tools for the analysis of whole genome sequencing projects (WGS500) with a special focus on immune disorders and cancers. He then joined the Department of Oncology to create a Bioinformatics group to both provide support to the department and lead independently funded research.

In 2014, Jean-Baptiste Cazier joined the University of Birmingham taking up the new chair of Bioinformatics to create the Centre for Computational Biology. This university-wide effort aims to promote excellence in Computational Biology, Data Science for the Life Sciences, and Bioinformatics across the range of fundamental and applied sciences, in both the University and allied Health Care arenas. In this post he has developed a Population Diversity approach and now leads the Population Diversity domain of the 100,000 Genome Project and extended his collaborations from Lebanon to Chilean, Brazilian, Indian and Chinese populations. He has been receiving funds from numerous grants internal, national such as MRC, BBSRC, Cancer Research UK, Wellcome Trust, Innovate UK, as well as international EU H2020, worth more than £14M.

Research interests

Prof Cazier leads the "Population Diversity at varying scales" project with The Alan Turing Institute. The ambitious goal of improving the Health of Individuals, in all their diversity, can be achieved by the confluence of the complementary fields of population genetics, epidemiology, socio-economics, clinical and environmental studies underpinned by the commonality of data sciences at various scales. Each of these domains, in isolation or in pair, aspires to improve the population well-being through a diversity of approaches. We are therefore proposing to define a novel framework to allow the characterisation, integration, comparison of the underlying structures found in the diversity of data, across data types, quantitative (discrete or continuous) or qualitative (ordinal or nominal), scale and studies. 

This project aims to define a novel mathematical framework to make use of existing datasets across studies, independently of their examined conditions, datatype, scale and location. With this novel approach, we expect to obtain a better understanding of the underlying relationship between datatypes in various context; thus enabling the integration and projection of diverse and sparse datasets in any context. In fine, understanding underlying data structure and homologies will improve analysis, streamline collection and minimise cost and noise. While there are on-going efforts to select features from complex biomedical datasets, we aim to go one step beyond, and explore the structure of the information rather than datasets themselves, allowing us to compare and transpose these findings from study to study.