Genomics-enabled breeding for sorghum improvement in sub-saharan Africa

Abstract

Sorghum (Sorghum bicolor, L. Moench) is a staple cereal food crop for millions of people in Sub-Saharan Africa and Asia. However, drought due to low and unpredictable rainfall decreases its productivity in semiarid regions. Understanding the genetic architecture of adaptive traits (drought tolerance, photoperiodic flowering time, and panicle architecture) of sorghum germplasm from breeding programs across West Africa could contribute to efficient molecular breeding. Breeding priorities in West African sorghum improvement programs seek to develop drought-adapted varieties with yield advantages, early and moderate maturity. However, field phenotyping for adaptation in early generations is difficult and there is limited technology to rapidly develop better-adapted varieties. This study aimed to dissect the genetic architecture of adaptive traits to develop high-throughput breeder-friendly markers for rapid introgression of adaptive alleles from donor to elites lines. In chapter 1, I describe the sorghum breeding programs in Senegal, the agronomic importance of sorghum types, and genomic approaches for crop improvement in semiarid regions. In chapter 2, I characterize 213,916 single nucleotide polymorphisms (SNPs) across 421 Senegalese sorghum accessions from the USDA-Germplasm Resources Information Network (GRIN) to identify genomic signatures of local adaptation. This study provided insights into the factors shaping the genetic diversity and the molecular systems underlying local adaptation to water scarcity in sorghum, a staple food security crop in Senegal. In chapter 3, I characterize 159,101 SNPs across 756 accessions of the West African sorghum association panel (WASAP) assembled from breeding programs of Senegal, Niger, Mali, and Togo. The genetic diversity structured by botanical types and subpopulations within botanical types across countries and large-effect quantitative trait loci (QTL) for photoperiodic flowering indicate an oligogenic architecture of flowering time in West African sorghum. In chapter 4, I use genome-wide SNP variation from chapter 3 and phenotypic data from multiple managed water stress environments to identify genomic regions associated with drought response. Significantly positive pleiotropic associations contributed to high phenotypic variance and colocalized with known stay-green (Stg) QTLs, suggesting the existence of Stg alleles in West African sorghum. Finally, in chapter 5, I summarize the expected steps to establish genomics-enabled breeding for sorghum improvement in West Africa. The genomic resources developed in this research have allowed for the dissection of the genetic architecture of adaptive traits. The SNPs associated with large-effect QTLs can be converted into high-throughput breeder-friendly markers for use in marker-assisted selection. These resources combined with discoveries from the global scientific community can be used to accelerate and facilitate the development of locally adapted varieties to meet global food demand in semiarid regions of Sub-Saharan Africa.