14 Oral - Platform Session 1 Complex Traits
Wednesday June 08, 11:35 AM - 11:50 AM

Authors:
Anastasia Teterina ^1,2; Peter Ralph ^1,3; Patrick Phillips ¹

Affiliations:
1) Institute of Ecology and Evolution, University of Oregon, Eugene, OR, USA; 2) Severtsov Institute of Ecology and Evolution RAS, Moscow, Russia; 3) Department of Mathematics, University of Oregon, Eugene, OR, USA

Keywords:
Theory & Method Development

Understanding how the genotype of an individual, through development, phenotype, and interaction with the environment, maps to its fitness is a fundamental question in evolutionary, quantitative, and systems biology. The structure of the gene regulatory networks (GRNs) that generate phenotypes can be shaped by both adaptive and non-adaptive forces. Many critical properties of GRNs, such as robustness against new mutations, genetic variability under stochastic and selective processes, evolvability of different topologies, have been studied using evolutionary simulations (Wagner 1996, 2008, 2012; Ibáñez-Marcelo & Alarcón 2014; Payne & Wagner 2014; Kioukis & Pavlidis 2018). To connect the evolutionary dynamics of GRNs with estimates and is possible to obtain empirically, we developed a simulation framework in SLiM3 (Haller & Messer 2016, 2019) with selection on gene expression levels in evolvable, dynamic gene regulatory networks encoded in realistically-sized genomes. Using the genomic data from such individual-based simulations, we estimated genetic diversity along the genome, conducted in silico molecular biological experiments such as estimation of the effects of knockout or overexpression of the gene on phenotypes, genetic variances, epistatic interactions, and described the structure and heterogeneity of evolved networks. We explored the correlations between genetic diversity, quantitative genetic measures, network centrality statistics, phenotypic effects of molecular manipulations on the genes, and strength of selection on genes. Then, we assessed how the evolutionary trajectories of genes depend on their roles in GRNs, and evaluated our ability to predict distributions of fitness effects and the trajectory of evolution based on network structure. Finally, we utilized neural networks to identify the key components of the GRNs that changed under directional and stabilizing selection.