Sign Epistasis and the Geometry of Interactions

TL;DR

This paper introduces a new framework combining sign epistasis and geometric interaction theories to analyze gene interactions, providing tools for large genetic systems and revealing prevalent order perturbations in malaria parasite fitness landscapes.

Contribution

It develops an integrated approach that enhances detection of sign epistasis and order perturbations, with practical applications to malaria genetics.

Findings

Order perturbations beyond sign epistasis are common in drug-free environments.

Reversed evolution to ancestral types is difficult due to prevalent order perturbations.

Rank orders of genotypes relate to additivity in fitness landscapes.

Abstract

Approaches to gene interactions based on sign epistasis have been highly influential in recent time. Sign epistasis is useful for relating local and global properties of fitness landscapes, as well as for analyzing evolutionary trajectories and constraints. The geometric theory of gene interactions, on the other hand, provides complete information on interactions in terms of minimal dependence relations. We propose a new framework that combines aspects of both approaches. In particular, we provide efficient tools for identifying sign epistasis and related order perturbations in large genetic systems, with applications to the malaria-causing parasite Plasmodium vivax. We found that order perturbations beyond sign epistasis are prevalent in the drug-free environment, which agrees well with the observation that reversed evolution back to the ancestral type is difficult. As a theoretical application, we investigate how rank orders of genotypes with respect to fitness relates to additivity.