Agent-based Model for Microbial Populations Exposed to Radiation (AMMPER) simulates yeast growth for deep-space experiments

Abstract

Space radiation poses a substantial health risk to humans traveling beyond Earths orbit to the Moon and Mars. As microbes come with us to space as model organisms for studying radiation effects, a computational model simulating those effects on microorganisms could enable us to better design and interpret those experiments. Here we present Agent-based Model for Microbial Populations Exposed to Radiation (AMMPER), which simulates the effects of protons, a major component of deep-space radiation, on budding yeast (Saccharomyces cerevisiae) growth. The model combines radiation track structure data from the RITRACKS package with novel algorithms for cell replication, motion, damage, and repair. We demonstrate that AMMPER qualitatively reproduces the effects of 150 MeV proton radiation on growth rate, but not lag time, of wild type and DNA repair mutant yeast strains. The variance in AMMPERs results is consistent with the variance in experimental results, suggesting that AMMPER can recapitulate the stochasticity of empirical experiments. Finally, we used AMMPER to predict responses to deep space radiation that may be tested in future experiments. A user-friendly, open-source, extendable Python package for studying the relationship between single-particle radiation events and population-level responses, AMMPER can facilitate the basic research necessary to ensure safe and sustainable exploration of deep space.