A Quantitative Modelling Approach for DNA Repair on a Population Scale

Abstract

AbstractDespite intensive research of DNA repair after UV in eukaryotes, a framework to quantitatively describe the dynamics in vivo is still lacking. We developed a new data-driven approach to analyse CPD repair kinetics over time in Saccharomyces cerevisiae. In contrast to other studies that consider sequencing signals as an average behaviour, we introduce a hidden axis representing independent cells where loci can transition from damaged to repaired. This permits the application of the Kolmogorov-Johnson–Mehl–Avrami model to find a region-specific and continuous representation of the entire temporal process. We correlated the parameters via a k-nearest neighbour approach to a variety of genomic features, including transcription rate and nucleosome density. The clearest link was found for the gene size, which has been unreported for budding yeast to our knowledge. The framework hence allows a comprehensive analysis of nuclear processes on a population scale.Author SummaryAs DNA encodes our very identity, it has been subject to a plethora of studies over the last century. The advent of new technologies that permit rapid sequencing of large DNA and RNA samples opened doors to before unknown mechanisms and interactions on a genomic scale. This led to an in-depth analysis of several nuclear processes, including transcription of genes and lesion repair. However, the applied protocols do mostly not allow a high temporal resolution. Quite the contrary, the experiments yield often only some few data signals over several hours. Missing dynamics between time points are chiefly ignored, implicitly assuming that they straightforwardly transition from one to another. Here, we show that such an understanding can be flawed. We use the repair process of UV-induced DNA damage as an example to present a quantitative analysis framework that permits the representation of the entire temporal process with only three parameters. We subsequently describe how they can be linked to other heterogeneous data sets. Consequently, we evaluate a correlation to the whole kinetic process rather than to a single time point. Although the approach is exemplified using DNA repair, it can be readily applied to any other mechanism and sequencing data that represents a state transition between two states, such as damaged and repaired.