University of Minnesota
Deploying high-throughput phenotyping for genomic assisted breeding: pathway to accelerated genetic gain and efficiency in potato
Abstract
dc:description.abstractPotato (Solanum tuberosum) is one of the most widely cultivated non-grain food crops in the world, playing a critical role in global food security. It is utilized as a vegetable, livestock feed, and a source of raw materials for various industrial products. However, yield improvement through traditional breeding approaches has stagnated. This is largely attributed to the crop’s genome complexity, characterized by its autotetraploid, clonal, and outcrossing nature. Despite recent advancements in ploidy theory and the growing availability of genomic resources, challenges in accurate phenotyping and the comprehensive evaluation of numerous traits persist. This research investigates innovative phenomic technologies to enhance genomic-assisted breeding in two potato market classes: chips and fresh market. The studies employ high-throughput phenotyping approaches, integrating digital imaging, drone-derived multispectral data, and machine learning techniques to address key challenges.The first study explored the use of quantitative phenotypic scores based on digital images for improving genomic selection of quality related traits. Analysis revealed significant contributions of additive and non-additive genetic effects, offering critical insights into the genetic mechanisms underlying these traits and their implications for potato breeding. The second study explored leveraging drone derived multispectral image data for phenomic selection and its potential to augment genomic selection. Highlight of this work shows phenomic selection surpassed genomic selection for yield prediction, with a combined phenomic-genomic approach improving prediction accuracy by over 30%. The third study assessed drone-derived multispectral variables to predict total plant nitrogen (N). Machine learning models achieved moderate-to-high accuracy in N prediction, with varying important features selected for improved prediction. The findings underscore the transformative potential of integrating phenomics with genomics to overcome breeding barriers in potato, improving the precision and efficiency of genetic enhancement strategies.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Yusuf, Muyideen
Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Yusuf, Muyideen. (2024). Deploying high-throughput phenotyping for genomic assisted breeding: pathway to accelerated genetic gain and efficiency in potato. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/270635.