Characterizing the structure and drivers of variation in behavior and gene expression

Abstract

Exploring the biological basis of inter-individual variation across different levels of biological organization is crucial for elucidating the underlying mechanisms that drive diversity in phenotypic expression, responses to environmental stress, and the evolution of complex traits. To address this problem, I follow two threads: characterizing complex behavioral patterns and characterizing the landscape of gene expression variation. In Chapter 1, we focus on quantifying complex behavioral traits. We explore how fine-scale behavioral changes occur over long timescales. To explore this, we developed methods for capturing continuous, high-resolution behavioral data of freely moving Drosophila melanogaster for up to 7 days. This unprecedented resolution underscores the importance of examining fine-scale behavior over extended timescales. Beyond the temporal context of behavior, in Chapter 3, we extended this framework to explore the effect of social context and group dynamics on individual behavior. To that end, we developed NAPS, which integrates pose estimation and tag-based identity assignment to allow simultaneous analysis of individual and group behaviors. We demonstrate its utility by studying social behavior in Bombus impatiens colonies. Our approach opens avenues for exploring individual behavioral variation within social contexts and its impact on group dynamics, including collective decision-making and social evolution. A key feature of these behavior-focused projects has been the striking degree of inter-individual variability. To explore factors associated with variation in variance, we turned to gene expression in Chapter 4. Transcriptional variance is not just noise but a biological trait that influences fitness, behavior, and evolution. While previous studies have focused on mean expression levels, we use large, publicly available gene expression data sets to establish that expression variance is consistent across tissues and data sets --- indicating a potential role for selection in its consistency. By correlating expression variance with genomic annotations and biological function, we aim to understand the underlying drivers and implications of gene expression variability. Finally, in Chapter 5, we introduce a novel, cost-effective strategy for validating SNP-level genotype-phenotype relationships in outbred Drosophila melanogaster. We use this method to confirm a specific SNP's effect on lifespan and outline how it can be applied to other organisms and traits.