|2021||Professor (W3 oL) of Genomic Microbiology and Immunity, Institute of Medical Microbiology and Hospital Hygiene, Heinrich Heine University Düsseldorf|
|2020-2021||Professor (W2) of Bioinformatics, Institute of Medical Statistics and Bioinformatics & CECAD, University of CologneUniversity of Cologne|
|2017-2021||Independent Manchot Group Leader, Institute of Medical Microbiology and Hospital Hygiene, Heinrich Heine University Düsseldorf|
|2016-2017||National Institutes of Health (NHGRI-NIH), US|
|2015||Co-founder Lighthouse Cancer Diagnostics Ltd, UK|
|2012-2014||Wellcome Trust Centre for Human Genetics, Oxford, UK|
|2008-2012||Department of Statistics, University of Oxford, UKp Leader, Institute of Medical Microbiology and Hospital Hygiene, Heinrich Heine University Düsseldorf|
- Genome Graphs now on biorxiv.
We work on translating advances in DNA sequencing technology into biological insight and novel diagnostic approaches. To achieve this we're using computational genomics and genome informatics; we develop new approaches for the analysis of sequencing data and apply these to large datasets (sometimes comprising tens or hundreds of thousands of individuals). In close collaboration with our colleagues, we're working on novel diagnostic approaches that leverage the power scalable DNA sequencing technology for the rapid and comprehensive interrogation of clinical samples.
Specific research interests include:
- Population Reference Graphs and variation-aware data analysis algorithms.
- Classical algorithms for the analysis of sequencing data assume that the genome of sequenced individual can be reconstructed by using a so-called 'reference genome' as a template.
- This approach fails in genomic regions in which the sequenced genome is strongly diverged from the reference genome, which is the case for many regions of high biomedical importance (e.g. the MHC and LRC/KIR regions of the human genome or those encoding the surface antigens in P. falciparum).
- Population Reference Graphs (PRGs), by contrast, include an explicit model of population sequence diversity - the sequenced genome is reconstructed from a recombining template of multiple genomes. We use PRGs to study the human immune system. More generally, we also work on algorithms for regions that undergo non-homologous combination, such as the LRC/KIR region and many bacterial sequences.
- Human immunogenetics and its role in disease architecture.
- We apply our tools, including HLA*IMP:02 and HLA*PRG, to large cohorts and characterize the loci and alleles that contribute to increasing or decreasing risk of specific diseases (e.g. for multiple sclerosis and psoriasis, but also for the UK Biobank. This leads to an improved understanding of disease etiology and can yield direct biological insights. We're also participating in projects that aim to improve our understanding of worldwide immunogenetic variation, in particular in African population, and its impact on health-related phenotypes.
- Algorithms for analysis of single-molecule sequencing data.
- We develop algorithms for the rapid analysis of single-molecule sequencing data, based on dimensionality reduction techniques from other fields of computer science. We're particularly interested in algorithms that scale to massive databases, such as RefSeq; algorithms that work well when applied to heterogeneous/metagenomic input samples; and algorithms specifically for the analysis of Nanopore sequencing data.
- Nanopore-based microbial diagnostics.
- We're working on real-time applications of Nanopore sequencing for the improved diagnosis of microbial infections. We're also interested in hospital outbreak tracking, epidemiology of mobile genetic elements, and the potential of sequencing to improve sepsis diagnosis.