Bioinformatics and Genomics
October 2018 Group Leader in the Bioinformatics and Genomics Programme at the Centre for Genomic Regulation, Barcelona, Spain.
2013-2018 Postdoctoral researcher at the Department of Medicine and Department of Biomedical Informatics, Harvard Medical School, Boston, USA
2012 PhD in Theoretical Physics, University of Cologne, Germany
Mutation, selection and stochasticity combine to leave their footprint in genetic data. With the proper tools and analyses, the resulting information can be leveraged to obtain estimates of selection, uncover mutational processes, and more intricate evolutionary dynamics. I am interested in a variety of topics in the field of genetics, with a strong focus on cancer as an evolutionary system and selection as a readout. However, analysis of human genetic data and mutational signatures will also remain of interest. The main focus of the group is the quantitative description of evolutionary processes, through the development of new probabilistic models and computational methods.
We are mainly interested in inferences at the cancer cell population level. This involves investigating different modes of selection active during tumorigenesis, including negative selection, in both the coding as well as the noncoding part of the genome. We analyze sequencing and other datasets to develop computational approaches to estimate selection. An important aspect of this is accounting for the heterogeneity of mutational signatures. Estimates of the strength of selection in cancer allow for a prioritization of genes and noncoding regions by their disease relevance, with the ultimate goal of promoting therapeutic advances. We also extend some of the analyses to the level of human population genetics.
We are looking for postdocs and PhD students with a background or strong interest in the quantitative description of evolutionary phenomena. This involves data analysis, mathematical modeling and development of computational approaches to address questions regarding cancer as well as human evolution. If you are interested, do not hesitate to write an email to Donate to discuss projects and possibilities!
Cancer Bayesian Selection Estimation (CBaSE)
We have developed a tool which derives gene-specific estimates of the strength of negative and positive selection in cancer. It accounts for the heterogeneity of mutation rate across the cancer genome, independent of external input information on mutation rate covariates. CBaSE estimates also take into account the context-dependent cancer type-specific mutational signature.
The CBaSE method and results were published in Weghorn & Sunyaev, Nature Genetics, 2017:
The CBaSE software can be downloaded as a standalone tool or used in a browser-based application here: