Unravelling the molecular mechanisms of genetic interactions: from yeast to cancer

Abstract

One of the hurdles to predict the phenotype of an individual based on their genotype is the presence of genetic interactions, also known as epistasis. A genetic interaction is a phenomenon in which the effect of the mutation of one gene depends on the presence of other mutations. The general goal of the work described in this thesis is aimed at understanding the mechanisms underlying genetic interactions, answering the question how do genetic interactions arise? The simplest example of a genetic interaction is a redundancy relationship between two genes. Redundancy occurs when one gene can take over the function of another gene, for example if they both code for highly similar proteins. In the case of a complete redundancy relationship, inactivation of either gene on its own will result in no detectable defect. Only simultaneous deletion of both genes will result in obvious defects. The role of a gene is often investigated by assessing functional consequences of their removal from the cell. Depending on the sensitivity of the assay used for assessing the effect of gene removal, between 66% and 53% of yeast gene deletions show no detectable defect when analyzed under a single condition. It is known that this non-responsive behaviour is caused by redundancy or condition dependency. We provide a systematic classification of the underlying causes of and their relative contribution to nonresponsive behavior upon gene deletion. Depending on the state or fate of a cell, different genes are expressed at different levels. This is in part mediated by gene-specific transcription factors (GSTFs). Understanding the basis of genetic interactions between GSTFs is therefore important for understanding the transcription regulatory network. We investigated genetic interactions between 72 pairs of GSTFs in yeast. This high-resolution expression atlas provides a systems-level overview of the genetic interaction landscape between GSTFs and reveals underlying mechanistic details. Besides revealing new redundancy relationships, this study also proposes two new molecular mechanisms of genetic interactions. We compared genetic interaction patterns identified using gene expression profiling for two classes of genes: gene specific transcription factors and signalling related genes. We employed exhaustive modelling to unravel putative molecular differences underlying different genetic interaction patterns. Our study proposes a new mechanistic explanation for a genetic interaction pattern that is more associated with gene specific transcription factors compared to signalling related genes. Overall, our findings and the methodologies implemented can be valuable for understanding the molecular mechanisms underlying genetic interactions. Finally, a framework for detecting genetic interactions in pediatric cancer is described. A map of genetic interactions is presented by combining the results obtained from four approaches. The results show a limited number of genetic interactions, in part probably due to the low number of samples. The results nevertheless also contain mutually exclusive mutations that have previously been reported, confirming the utility of the approach. Our framework therefore provides a basis for future detection of genetic interactions between different types of mutations and in larger datasets.