High-Throughput Phenotyping and Random Regression Models Reveal Temporal Genetic Control of Soybean Biomass Production

• DOI: 10.3389/fpls.2021.715983
• Publication Date: 2021-09-04T10:41:05Z
• Authors: Fabiana Freitas M...; Hinayah Rojas de ...; Miguel Angel Lopez; Bilal Jamal Abughali; Guilherme Gomes; Keith Aric Cherkauer; Luiz Fernando Brito; Katy Martin Rainey
• Keywords: Genetic architecture; Genomic Selection

Understanding temporal accumulation of soybean above-ground biomass (AGB) has the potential to contribute yield gains and development stress-resilient cultivars. Our main objectives were develop a high-throughput phenotyping method predict AGB over time reveal its quantitative genomic properties. A subset SoyNAM population ( n = 383) was grown in multi-environment trials destructive measurements collected along with multispectral RGB imaging from 27 83 days after planting (DAP). We used machine-learning methods for phenotypic prediction AGB, breeding values, genome-wide association studies (GWAS) based on random regression models (RRM). RRM enable study changes genetic variability further allow selection individuals when aiming alter general response shapes time. predictions high R 2 0.92–0.94). Narrow-sense heritabilities estimated ranged low moderate (from 0.02 at 44 DAP 0.28 33 DAP). adjacent had highest correlations compared those apart. observed accuracies biases indicating that values can be predicted specific intervals. Genomic regions associated varied time, no markers significant all points evaluated. Thus, seem powerful tool modeling architecture provide useful information crop improvement. This provides basis future combine analyses understand complex longitudinal traits plants.