agile genetics: single gene resolution without the fuss

Authors: Vaughn, Justin; KORANI, WALID - Hudsonalpha Institute For Biotechnology; CLEVENGER, JOSH - Hudsonalpha Institute For Biotechnology; OZIAS-AKINS, PEGGY - University Of Georgia

Submitted to: Bioessays
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/8/2024
Publication Date: 5/20/2024
Citation: Vaughn, J.N., Korani, W., Clevenger, J., Ozias-Akins, P. 2024. agile genetics: single gene resolution without the fuss. Bioessays. https://doi.org/10.1002/bies.202300206.
DOI: https://doi.org/10.1002/bies.202300206

Interpretive Summary: Gene discovery reveals new biology and expands the utility of marker assisted selection for crop improvement. Such knowledge also enables targeted mutagenesis for trait engineering. Still, these discoveries can take over a decade, particularly if the gene is rare in the population. Based on the synthesis of our work and many others, we propose a general strategy – Agile Genetics. The framework is based on using natural mutations identified in cultivated and wild varieties. Experimental populations are then developed to interrogate this variation. Graph-based pangenomics involving all "founder" varieties of the population maximize accuracy and improve interpretation of the results. In conjunction with these bioinformatic innovations, large comparisons across extreme tails of traits expressed in the progeny allow genes to be identified to a very narrow range. Such large samples are enabled by avoiding individual sample replication and thus tagging and tracking. Based on this merger of agronomic scale with molecular and bioinformatic innovation, we predict a new age of rapid gene discovery.

Technical Abstract: Gene discovery dramatically expands the utility of marker assisted selection and allows for targeted mutagenesis. Here we review and synthesize current efforts to identify single genes within the context of a single, <3-year experimental framework that we summarize as "Agile Genetics". QTL-seq combines bulk segregant analysis with next generation sequencing to identify QTL in segregating populations and has the potential to resolve causal genes and even variants. The theoretical underpinnings for QTL-seq resolution for single-gene traits have been defined previously, yet the behavior of multi-gene traits with partial heritability and variable effect-sizes remains mathematically intractable. To overcome these hurdles, we simulated 729 combinations of interacting parameters using highly realistic genetic model derived directly from population data. At a population size of >5,000, single gene resolution can be achieved even in the context of multiple-gene traits. At this scale, read depth and technical replication become major drivers of resolution. To that end, we propose an iterative depth sequencing step that makes such requirements feasible, and we demonstrate its technical possibility. We also review and discuss major pillars of the Agile Genetics method, including exhaustive genotyping via pangenomic graphs and the importance of a nested sample within the larger bulked material. Finally, we present an interactive web-tool, QTLSurge, to facilitate QTL-seq experimental design.