Dissection of genomic drivers of spike morphology changes in wheat by high-throughput phenotyping

Abstract

Spike morphology is crucial for wheat (Triticum aestivum L.) yield and environmental adaptation. We developed a high-throughput phenotyping platform to dissect spike morphology traits based on 54 traits in 1,359 wheat accessions. These 54 spike morphology traits exhibited clear geographical differences among 306 worldwide accessions and breeding selection trend across different time windows for 1,053 accessions released from 1900 to 2020 in China. Based on geographical distribution and breeding selection of haplotypes, we attribute the differences in spike morphology to variable haplotype combinations. Wheat breeding breaks the trade-off between spike length and width/thickness, resulting in increased spike volume. A large proportion of genomic regions has been identified across wheat varieties and utilized as a fixed group to facilitate the targeted improvement and selection of desirable traits during wheat breeding programs. Overall, we provide a resource for the molecular design of spike morphology to facilitate future wheat breeding.

Code and data availability

The paper's high-resolution spike phenotyping platform software is explicitly released as public code by the authors on GitHub. The genotype datasets (GVM000272/GVM000720) are molecular omics deposits and do not qualify as phenotype/trait data; other listed tools are generic third-party libraries.