Dynamic genetic architecture of yeast response to environmental perturbation shed light on origin of cryptic genetic variation

Abstract

Cryptic genetic variation could arise from, for example, Gene-by-Gene (G-by-G) or Geneby-Environment (G-by-E) interactions. The underlying molecular mechanisms and how they influence allelic effects and the genetic variance of complex traits is largely unclear. Here, we empirically explored the role of environmentally influenced epistasis on the suppression and release of cryptic variation by reanalysing a dataset of 4,390 haploid yeast segregants phenotyped on 20 different media. The focus was on 130 epistatic loci, each contributing to segregant growth in at least one environment and that together explained most (69-100%) of the narrow sense heritability of growth in the individual environments. We revealed that the epistatic growth network reorganised upon environmental changes to alter the estimated marginal (additive) effects of the individual loci, how multi-locus interactions contributed to individual segregant growth and the level of expressed genetic variance in growth. The estimated additive effects varied most across environments for loci that were highly interactive network hubs in some environments but had few or no interactors in other environments, resulting in changes in total genetic variance across environments. This environmentally dependent epistasis was thus an important mechanism for the suppression and release of cryptic variation in this population. Our findings increase the understanding of the complex genetic mechanisms leading to cryptic variation in populations, providing a basis for future studies on the genetic maintenance of trait robustness and development of genetic models for studying and predicting selection responses for quantitative traits in breeding and evolution.

Dissecting the Genetic Regulation of Yeast Growth Plasticity in Response to Environmental Changes

“…In our previous analysis of this dataset [18,19], 11 epistatic networks with 13 hubs were detected underlying the variation of growth means. These epistatic networks were highly environmental-dependent, with the radial loci being connected or disconnected in response to changing the growth medium.…”

“…Growth data from 2 of the 20 media (YPD and YNB) were excluded from the analyses here since they were used as control media to normalize the environmental effect for the remaining 18 media. Previously detected additive QTL, epistatic QTLs, as well as within-environment interaction networks, were downloaded from the supplementary data of Forsberg et al [19] and Zan et al [18].…”

“…Here, we continue the analyses of an experimental yeast dataset where we earlier identified extensive genetic epistatic interaction networks. These influenced yeast growth and had variable effects across the multiple evaluated growth media (G-by-G-by-E interactions) [18,19]. A remaining question from that study was the potential link between the changes in the genetic interaction networks across environments (G-by-G-by-E) and the phenotype plasticity of individual segregants (genotypes).…”

Gene–Environment Interaction in the Era of Precision Medicine – Filling the Potholes Rather Than Starting to Build a New Road

“…In other words, since the machinery here provided has proven useful to fill in inconvenient potholes of the classical road network, it should be deemed to be applicable to increasingly complex challenges in the future. For instance, the need to consider gene–gene–environment interaction (e.g., Zan and Carlborg, 2020 ) and/or gene–environment–environment interaction (e.g., Keers and Pluess, 2017 ) has already arisen and it is to this regard worth highlighting here that the advantages of ARNOIA can also be applied to address such complexities (and actually gene–environment interaction/correlation, multiple alleles, dominance, epistasis and departures from Hardy-Weinberg equilibrium and from linkage equilibrium, simultaneously) by merging the mathematical developments provided above with previous theory ( Álvarez-Castro and Yang, 2011 ; Alvarez-Castro and Crujeiras, 2019 ).…”

A genotype-phenotype-fitness map reveals local modularity and global pleiotropy of adaptation

“…For example, perhaps these expression changes are coordinated and reflect the up-regulation of a stress-response pathway. Unfortunately, defining the functional units in which a gene product participates remains difficult, especially because these units re-wire across genetic backgrounds, environments, and species (Geiler-Samerotte et al, 2019;Pavličev et al, 2017;Sun et al, 2020;Zan and Carlborg, 2020).…”