Project Team
- Dr. Alexander Lipka, Assistant Professor, Dept. of Crop Sciences
- Dr. Lindsay Clark, Research Specialist, Dept. of Crop Sciences
- Dr. Aurelie Lozano, Research Staff Member, IBM Research
- Dr. Naoki Abe, Distinguished Research Staff Member, IBM Research
Abstract
The potential for genome-wide association studies (GWAS) to identify and quantify the effect sizes of genomic loci contributing to quantitative trait variability has not been fully realized in part because the statistical approaches typically employed oversimplify the contributions of multiple loci affecting multiple phenotypes. To address this, Co-PI Lozano and collaborators have incorporated machine learning and regularization into a GWAS model, thus making it possible to quantify the contributions of multiple loci to multiple phenotypes in a single statistical analysis. In this proposal, we seek to augment this novel GWAS model so that it can be used in a wider variety of agronomical data, and then generalize the regularization approach so that the potential of this GWAS model to elucidate the genetic architecture of complex traits is maximized. Finally, we will leverage the expertise of Co-PIs Lipka and Clark to assess the performance of the GWAS models we develop through evaluation of simulated and real phenotypic data from three agronomically important crops. This proposed research will establish a solid foundation of model development, refinement, and validation, which in turn will significantly strengthen future proposals for federally funded research, where we will seek to apply these GWAS models to emerging “big data” in agriculture.