Research Assistant Professor Michigan State University
The lack of a visible mutant phenotype is often attributed to redundancy between genes, especially in plants, which have large numbers of gene duplicates derived from repeated rounds of genome duplication. However, there are several alternative explanations for redundancy, including ongoing functional decay, conditional effects, and phenotypic effects that although subtle may nevertheless affect fitness. Thus, the degree of genetic redundancy is likely overestimated, and this hampers efforts to understand duplicate gene functions and evolution. We hypothesize that lifetime fitness is a better measure of the degree of selection on a gene than, for example, the presence of a morphological or biochemical phenotype, because it reflects the ability of a plant to survive and reproduce. We are measuring the fitness of single and double mutants of pairs of Arabidopsis thaliana gene duplicates derived from the most recent whole genome duplication event. Analysis of fitness data in the form of fruit counts have been recorded for 100 duplicate pairs. Because seed counts are a more accurate measure of fitness, we established a deep learning-based pipeline to efficiently count seeds. These data will be incorporated into machine learning models for predicting functional redundancy between Arabidopsis duplicate genes using features such as evolutionary conservation and correlated expression patterns. The fitness data will provide a new estimate of functional redundancy between genes and, combined with machine learning predictions, will provide insights into what the degree of genetic redundancy is between Arabidopsis genes and a clearer picture of how duplicate gene functions have evolved.
Coauthors: Jeffrey Conner – Michigan State University;Traverse Cottrell – Michigan State University;Siobhan Cusack – Michigan State University;Paityn Donaldson – Michigan State University;Sarah Horan – Michigan State University;Patrick Krysan – University of Wisconsin-Madison;Fanrui Meng – Michigan State University;Shin-Han Shiu – Michigan State University
