Open AccessImputation accuracy of wheat genotyping-by-sequencing (GBS) data using barley and wheat genome references
Author(s) -
Hadi Alipour,
Guihua Bai,
Guorong Zhang,
Mohammad Reza Bihamta,
Valiollah Mohammadi,
S A Peyghambari
Publication year - 2019
Publication title -
plos one
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.99
H-Index - 332
ISSN - 1932-6203
DOI - 10.1371/journal.pone.0208614
Subject(s) - imputation (statistics) , refseq , minor allele frequency , single nucleotide polymorphism , missing data , reference genome , genotyping , biology , genome , 1000 genomes project , genome wide association study , statistics , genetics , computational biology , data mining , computer science , mathematics , genotype , gene
Genotyping-by-sequencing (GBS) provides high SNP coverage and has recently emerged as a popular technology for genetic and breeding applications in bread wheat ( Triticum aestivum L.) and many other plant species. Although GBS can discover millions of SNPs, a high rate of missing data is a major concern for many applications. Accurate imputation of those missing data can significantly improve the utility of GBS data. This study compared imputation accuracies among four genome references including three wheat references (Chinese Spring survey sequence, W7984, and IWGSC RefSeq v1.0) and one barley reference genome by comparing imputed data derived from low-depth sequencing to actual data from high-depth sequencing. After imputation, the average number of imputed data points was the highest in the B genome (~48.99%). The D genome had the lowest imputed data points (~15.02%) but the highest imputation accuracy. Among the four reference genomes, IWGSC RefSeq v1.0 reference provided the most imputed data points, but the lowest imputation accuracy for the SNPs with < 10% minor allele frequency (MAF). The W7984 reference, however, provided the highest imputation accuracy for the SNPs with < 10% MAF.