Hybrid performance as related to genomic diversity and population structure in public sorghum inbred lines

Abstract

AbstractThe low‐cost next‐generation sequencing technologies provide tremendous opportunities for dissecting complex traits in crop species. The objective of this study was to assess the relationship between genomic diversity, population structure, and hybrid performance in public sorghum [Sorghum bicolor (L.) Moench] inbred lines. A total of 279 public sorghum inbred lines (228 R‐lines and 51 B‐lines) developed across more than two decades were studied. The inbreds were genotyped using genotyping‐by‐sequencing (GBS) platform, which generated 282,536 single nucleotide polymorphisms (SNPs). After filtering at a ≤5% threshold for minor allele frequency (MAF) and <20% missing data, a total of 66,265 SNPs were returned and used for analysis. Mean polymorphic information content (PIC) was 0.35, and gene diversity across the inbreds was 0.46. The neighbor‐joining tree, principal component, and STRUCTURE analyses clustered the inbreds into three subgroups. One‐hundred and two test‐cross hybrids, 50 between closely related parents and 52 between distantly related parents, were developed and evaluated along with two commercial checks in two environments using three replications. Data were obtained on plant height, maturity, yield, and yield components. Mean performance of hybrids derived from closely related and distantly related parents was compared to determine the value of genomics‐based genetic distance to predict hybrid performance. The results revealed the presence of robust genetic variability and hierarchical genetic structure among inbred parents, but marker‐based genetic distance was not a good predictor of yield performance.