Parent-progeny imputation from pooled samples for cost-efficient genotyping in plant breeding

Area of Science:

• Plant breeding and genetics
• Genomics and bioinformatics
• Quantitative genetics

Background:

• Whole-genome selection (WGS) and molecular breeding methods require high-throughput genotyping of numerous individuals.
• Biparental populations of homozygous individuals are common in plant breeding, facilitating parent-progeny imputation.
• Current imputation methods may not be cost-effective for very large populations.

Purpose of the Study:

• To develop and validate a novel parent-progeny imputation method for inferring individual genotypes from non-barcoded pooled DNA samples.
• To assess the accuracy and reliability of genotype imputation using a Hidden Markov Model (HMM).
• To evaluate the impact of imputation on genomic estimated breeding values (GEBV) and its potential for cost reduction in genotyping.

Main Methods:

• Development of a Hidden Markov Model (HMM) for parent-progeny imputation from pooled DNA samples.
• Simulation of maize double haploid (DH) populations genotyped by sequencing (GBS) at varying coverage levels (0.125x to 4x) for 3,000 loci.
• Comparison of imputed marker scores and GEBV with true marker scores and GEBV.

Main Results:

• High concordance observed between true and imputed marker scores.
• HMM provided well-calibrated genotype probabilities, accurately reflecting imputation uncertainty.
• GEBV calculated from imputed scores closely matched those from true marker scores, with within-population correlations approaching 0.95 at 1x and 4x coverage.
• The method demonstrated potential to reduce genotyping costs by a factor up to the number of pooled individuals.

Conclusions:

• The developed HMM-based imputation method enables accurate and cost-effective genotyping from pooled DNA samples in plant breeding.
• This approach facilitates large-scale genotyping for applications like whole-genome selection (WGS) without increasing sequencing costs.
• The method supports reliable estimation of genomic breeding values, crucial for accelerating genetic gain in breeding programs.