Location: Sunflower and Plant Biology Research
Project Number: 3060-21000-043-19-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Sep 1, 2018
End Date: Aug 31, 2023
The goal of this project is to provide technicians, students, and private sector plant breeders with a simplified module written in R or Python that provides prediction statistics given a phenotypic and genotypic database for sunflower lines and hybrids.
A NDSU graduate student in the Plant Sciences Department will design a small Python or R-based Graphical User Interface to integrate our MySQL database query functions with R-scripts and Python functions that have already been developed for genomic selection models. Specifically, we will use R packages that can handle GCA and SCA effects (multiple random effects) such as ‘BGLR’ for Bayesian GS models and ‘sommer’ for GBLUP. Cross-language incompatibility and non-uniformity of parameters in function calls are the complication here, but frameworks, such as RPy2, can translate between environments (RPy2 Working Group, 2017). The work will involve working closely with USDA sunflower breeding program personnel to make a design that is useful to people not well versed in programming. The student, who has already been hired, has training in computer science and plant breeding and will use best practices to develop the program.