Implementation of Genomic and Phenomic Tools for Introgression of Reinstated Sorghum Conversion (RSC) Germplasm

Genotypic variance is necessary for trait improvement as limited diversity can reduce genetic gain in crop improvement. To maintain genetic diversity, a wealth of germplasm exists in the USDA-ARS sorghum [Sorghum bicolor (L.) Moench] collection, but most of the accessions are not adapted to temperate climates. Methodologies aimed at incorporating tropical germplasm have been evaluated extensively by public and private breeding programs due to their beneficial alleles for improved agronomic performance. However, concerns as to how and when material from this program should be tested for its agronomic value have been expressed. Three different methodologies were conducted to assess partially converted, early-generation lines from the Reinstated Sorghum Conversion (RSC) program. Our first methodology was to investigate the utility of using markers for the identification of high levels of tropical genome recovery, while elucidating the relationship between marker data and agronomic performance. The utilization of markers to predict hybrid performance was not observed, nonetheless, the ability to prescreen lines with high amounts of tropical genome recovery proved useful. Expanding upon these results, the second methodology focused on the phenotypic evaluation of partially converted, early-generation lines. From the lines evaluated, I was able to release lines that combined agronomic productivity with greater genetic diversity as confirmed via genotyping-by-sequencing. These eleven parental germplasms are being released to provide new genetic diversity for forage and grain hybrid improvement programs. Finally, noticing the value of phenotypic observations and its implications on selecting valuable germplasm, I further investigated plant height using high-throughput phenotyping via unmanned aerial systems (UAS). Within both advanced and early generation sorghum trials, genotypic variation estimates were comparable to manual measurements with highly repeatable estimates of plant height, indicating the value of UAS in plant breeding programs