Inferring modularity in genotype-phenotype maps using patterns of epistasis and pleiotropy
Methods for inferring the relationship between genotype and phenotype often search for low-dimensional structure, motivated by the hypothesis that genetic loci can be organized into a small number of groups representing biological processes. These methods typically group genetic loci based on which other loci they interact with (epistasis) or based on how their effects covary across environmental…
Researchers have developed two methods to uncover the modular structure of genotype-phenotype maps by integrating patterns of epistasis and pleiotropy. Epistasis refers to the interaction between genetic loci, while pleiotropy describes how a single gene influences multiple traits.
The first approach, called Sparse, Low-dimensional Epistatic Decomposition (SLED), models the genotype-phenotype relationship using a small number of modules. Each gene contributes to these modules through sparse, additive, and epistatic effects. Modules, in turn, contribute to multiple phenotypes in a similar manner.
The second method, Graph Tensor Decomposition (GTD), represents epistatic interactions in different environments as a combination of shared basis graphs. Each graph captures a recurring pattern of epistasis across environments. GTD does not rely on additive effects, allowing for more flexibility in the learned graphs.
To test the effectiveness of these methods, the researchers applied them to experimental data from yeast segregants grown in 101 different environments. This structure mimicked a hub-and-spoke network, with a central environment connecting to others.
The researchers evaluated how well the inferred modules aligned with the biological structure of the data. They discovered that both SLED and GTD produced modules that generalized well to previously unseen environments. Furthermore, these modules were highly similar to those obtained using an existing method that focused solely on additive effects.
These findings suggest that the genetic architecture underlying the yeast dataset is consistent across different aspects of the genotype-phenotype map, including additive effects, epistatic effects, and their joint influence. The researchers' methods offer a comprehensive framework for integrating epistasis and pleiotropy to uncover interpretable low-dimensional structure in genotype-phenotype maps, building upon previous work that often treated these factors independently.
Written by urgent.news from bioRxiv's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
