Genetic Diversity and Genome-Wide Association Studies in Sorghum (Sorghum bicolor (L) Moench) Genotypes for Drought Adaptation in Ethiopia

Abstract

Ethiopia, as a center of origin and diversity for sorghum, holds extensive collections that have been identified as valuable sources of genes to address production challenges and enhance yield; however, detailed genetic characterization and effective utilization in breeding programs remain limited. Moreover, sorghum productivity in Ethiopia is below the global average due to various biotic and abiotic constraints, with drought being a major one. This study aimed to evaluate the genetic diversity of the Ethiopian sorghum [Sorghum bicolor (L.) Moench] genotypes and identify genomic regions linked to drought responsive seedling traits using SNPs markers. A total of 158 sorghum genotypes from various regions of Ethiopia were assessed for fifteen phenotypic traits under greenhouse conditions using a high-throughput phenotyping platform. Significant variations (p < 0.001) were observed for all traits, with strong and positive correlations among root traits (r = 0.76-0.99, p < 0.001) and shoot traits (r = 0.27- 0.98, p < 0.001). Cluster analysis based on the 15 traits grouped the genotypes into four genetic groups. Furthermore, 40 genotypes (20 with narrow root angles and 20 with wide root angles) were evaluated for growth and transpiration responses under well-watered (WW) and water stress (WS) conditions in a greenhouse. Significant differences (p < 0.01) in growth and transpiration responses were observed among the genotypes under both WW and WS conditions. Under WS conditions, shoot dry mass, root dry mass, total transpiration, and total water extraction were significantly lower (p < 0.05) in wide root-angled genotypes compared to narrow root-angled genotypes, while the mean difference in fraction of transpirable soil water (FTSW) was lower in narrow root-angled genotypes. Conversely, these attributes were higher under WW conditions, except for root growth and total water extraction, which showed non-significant variation (p > 0.05) between narrow and wide root angled genotypes. Genotypic data were used in a genome-wide association study (GWAS) to identify significant single nucleotide polymorphism (SNP) markers associated with 13 phenotypic traits. The GWAS, conducted using the “mrMLM.GUI” R package, employed 127,804 high-quality SNP markers. A total of 49 stable quantitative trait nucleotides were consistently identified by multiple models. Candidate genes associated with root and shoot traits included ALUMINUM ACTIVATED MALATE TRANSPORTER (ALMT), F-BOX domain, FAR1, RING ZINC FINGER, CHLORIDE CHANNEL (CLC), WRKY domain, and DEAD-box RNA helicase. Population structure analysis categorized the genotypes into five sub-populations. The genetic diversity of the sorghum collection was also assessed, with gene diversity (He) values ranging from 0.30 to 0.63 (average 0.51) and observed heterozygosity (Ho) values from 0.02 to 0.81 (average 0.13). PIC values ranged from 0.06 to 0.50 (average 0.32), and genetic distance varied from 0.13 to 0.95 (average 0.69), indicating high genetic distance among genotypes.Cluster and PCoA analyses showed clustering irrespective of regions of origin. AMOVA revealed 99% genetic variation within structured populations and regions. Fixation index (Fst) values (0.006 and 0.008) indicated low genetic differentiation among regional and structure sorghum populations. In conclusion, this study highlights the genetic basis of seedling root and shoot traits, supporting breeding strategies for drought-resilient sorghum. Further investigation of genetic mechanisms and candidate genes can advance the development of cultivars adapted to drought conditions.