< Back to previous page


Breast cancer-associated risks of established susceptibility loci in other cancers.

As a PhD student, I studied the potential of whole-genome sequencing data to assess various types of genetic variability (i.e. point mutations, insertions and deletions, copy number alterations, fusion genes, expression and methylation changes) and studied how each of them could contribute to a specific human disease. For this purpose, I developed, used, combined and adapted numerous different bioinformatical approaches and applied them to several whole-genome sequencing datasets (e.g., Complete Genomics and Illumina) in order to answer biological and clinical relevant questions. In particular, I studied sequencing data of familial amyotrophic lateral sclerosis (ALS), mismatch repair (MMR)-deficient cancers, uterine leiomyosarcoma (uLMS) and lung tumors.

First, I demonstrated that depth of coverage from WGS can be used to reveal repeat expansions. In combination with a tagging variant in a shared haplotype, this method confirmed the involvement of the (GGGGCC)n repeat expansions in C9orf72 as a cause of familial ALS.

In a second project, I performed whole-genome and whole-exome sequencing of MMR-deficient tumours. These tumours are characterized by very high mutation rates, requiring optimized variant filtering techniques to increase the number of true-positive and reduce false-negative variants. Furthermore, I demonstrated that MMR-deficient tumors carry a mutation signature distinct from other tumor genomes, but similar to germline DNA. This suggests that human genetic variation arises through mutations escaping MMR. Furthermore, I developed an MSI test based on these WGS data to detect MMR-deficient tumors. This MSI test has been patented and is being further developed by Biocartis into a clinical diagnostic test.

After sequencing uLMS cancer patients, characterized by genetic heterogeneity and a complex karyotype, I developed a novel method to uncover tumor driver genes. This model was successful in confirming the known involvement of the tumor suppressor gene RB1 and PTEN-pathway genes in uLMS. Furthermore by analysing copy number profiles of additional uLMSs, I identified additional genomic regions of prognostic relevance, i.e., deletion of 5q15, 7q36.3 (VIPR2) and 9q34.13.

Finally, I embarked on a computationally challenging project, in which I developed a strategy to integrate DNA methylation and gene expression data obtained from genome wide 450K Illumina methylation chips and transcriptome sequencing data. In this project, I studied the link between two respiratory diseases: COPD and lung cancer. This revealed that lung tumors that develop in a COPD environment differ from those that develop in patients without COPD. In particular, DNA methylation, gene expression and histological data all pointed to an altered anti-tumor immune response in COPD patients. Given the impaired immune response present in COPD lungs, these epigenetic events provide a biologically plausible and relevant link to lung cancer in COPD patients.

Date:27 Sep 2010 →  19 Nov 2015
Keywords:Consortia, Cancer risk, Genotype-environment interaction, Genotype-genotype interaction, Polymorphism, Bayesian network, Cancer, Breast cancer
Disciplines:Laboratory medicine, Palliative care and end-of-life care, Regenerative medicine, Other basic sciences, Other health sciences, Nursing, Other paramedical sciences, Other translational sciences, Other medical and health sciences
Project type:PhD project