IISER MOHALI

Abstract: Epistasis is a confounding phenomenon that exhibits several statistical patterns. Depending on the context, it can constrain or facilitate adaptation. Despite a volume of empirical evidence of epistasis, its mechanistic origin remains unknown. In this talk, I present a metabolic cell growth model that explains how global statistical epistatic patterns emerge from a simple resource allocation problem in a cell. It also explains the ubiquity of epistasis between promoters and genes, and predicts their inter-dependent adaptive trajectories. I will then present the analysis of a recent intramolecular fitness landscape in E. coli, and highlight two novel aspects of how epistasis acts. First, mutations exhibit one of two states - a small fraction of mutations exhibit strong global epistasis patterns. A large majority does not. Thus, epistasis is ‘binary’. Second, epistasis is also ‘fluid’ - the nature of epistatic interactions between two mutations is strongly contingent on the genetic background. Last, I will discuss the implications of the two studies on our understanding of epistasis and adaptation.