Wild perennial plants can contribute to improving the functionality of agricultural systems; however, these species are low-yielding, difficult to manage, and challenging to improve via classical…

The Department of Agriculture National Institute of Food and Agriculture awarded $285,000 to HudsonAlpha Institute for Biotechnology on April 1, 2026, under the Agriculture and Food Research Initiative (AFRI, CFDA 10.310) to develop genomic tools for domestication of wild perennial oilseed crops, specifically Silphium integrifolium and Silphium perfoliatum.

HudsonAlpha will deploy the Silphium50K genotyping platform—a universal high-throughput genotyping tool targeting 50,000 loci across both Silphium species—to genotype multiple populations and test genomic prediction models for seed weight, morphology, and oil content. Phenotyping will employ multispectral seed imaging via videometer and high-performance liquid chromatography to measure fatty acid composition. The genotypic and phenotypic data will be integrated to develop and validate prediction models for selecting superior individuals within a broader Silphium domestication program. The approaches are intended to reduce molecular marker and phenotype generation costs and timelines while increasing genetic gain per cycle, with applicability to other nascent crops with limited resources.

Performance takes place in Huntsville, Alabama, with a period of performance from April 1, 2026, through March 31, 2028. Saint Louis University serves as a sub-recipient, contributing development of universal high-throughput genotyping and seed phenotyping tools for Silphium integrifolium domestication. This is a project grant, a standard assistance type under AFRI.

Wild perennial plants can contribute to improving the functionality of agricultural systems; however, these species are low-yielding, difficult to manage, and challenging to improve via classical breeding. We recently completed a genome assembly for silphium integrifolium and silphium perfoliatum, two perennial species endemicto the united states that exhibit drought tolerance and produce seeds with high oil content. Given the large genome size (7.6 GB), which is predominantly composed of repetitive elements (~75%), we developed and tested a pipeline for target sequencing to enhance efficiency and reduce genotyping costs. A pilot test targeting 50,000 loci resulted in more than 90% of targets being sequenced and a median depth in each locus of 22.8 x. This new platform (SILPHIUM50K) includes all the genes annotated in both genomes and represents a universal genotyping platform for the entire silphium genus. In this project, we aim to utilize the SILPHIUM50K to genotype several populations, focusing on the deployment of genomic prediction to improve seed weight, morphology, and oil content. For phenotyping, multispectral images of seeds will be captured using a videometer, and the fatty acid composition will be measured using high-performance liquid chromatography. The phenotypes, in combination with the genotypic data, will be used to test prediction models, which will then be deployed for selecting individuals throughout the silphium domestication program. The approaches proposed here are expected to be applicable to other nascent crops with limited resources, reducing the costs and time required to generate molecular markers and phenotypes, and increasing the genetic gain per cycle.