Network properties influence covariance between gene expression and fecundity fitness of Caenorhabditis elegans in a novel laboratory environment

Gene expression bridges genotype and phenotype, shaping how organisms respond to their environments. Yet, it remains difficult to untangle how the architecture of gene regulatory networks (GRNs) influences the relationship between gene expression and fitness-especially in structured populations where genetic variants affecting both types of traits are inherited together. Here, we use publicly available data from wild Caenorhabditis elegans strains to link genome-wide gene expression levels to lifetime fecundity upon cultivation in a novel laboratory environment. Despite strong population structure caused by self-fertilization, selective sweeps, and hyperdivergent haplotypes, we find that a small subset of genes show significant covariance between their expression levels and fecundity fitness in laboratory cultivation conditions. These associations persist even after controlling for underlying genomic features. Genes that are older, show more tissue-specific expression patterns, and are particularly enriched in the germline and nervous system exhibit stronger covariance with fecundity in the new laboratory environment, consistent with known patterns of genetic divergence between wild and "domesticated" laboratory strains. Moreover, genes that are centrally positioned within GRNs, or are regulated downstream of certain transcription factors, show stronger associations with fitness. Together, these results suggest that genome structure and network topology jointly shape how variation in gene expression translates into fitness, shedding light on the early stages of adaptation to novel environments.