Effect of population size and unbalanced data sets on QTL detection using genome-wide association mapping in barley breeding germplasm

Abstract

Over the past two decades many quantitative trait loci (QTL) have been detected; however, very few have been incorporated into breeding programs. The recent development of genome-wide association studies (GWAS) in plants provides the opportunity to detect QTL in germplasm collections such as unstructured populations from breeding programs. The overall goal of the barley Coordinated Agricultural Project was to conduct GWAS with the intent to couple QTL detection and breeding. The basic idea is that breeding programs generate a vast amount of phenotypic data and combined with cheap genotyping it should be possible to use GWAS to detect QTL that would be immediately accessible and used by breeding programs. There are several constraints to using breeding program-derived phenotype data for conducting GWAS namely: limited population size and unbalanced data sets. We chose the highly heritable trait heading date to study these two variables. We examined 766 spring barley breeding lines (panel #1) grown in balanced trials and a subset of 384 spring barley breeding lines (panel #2) grown in balanced and unbalanced trials. In panel #1, we detected three major QTL for heading date that have been detected in previous bi-parental mapping studies. Simulation studies showed that population sizes greater than 384 individuals are required to consistently detect QTL. We also showed that unbalanced data sets from panel #2 can be used to detect the three major QTL. However, unbalanced data sets resulted in an increase in the false-positive rate. Interestingly, one-step analysis performed better than two-step analysis in reducing the false-positive rate. The results of this work show that it is possible to use phenotypic data from breeding programs to detect QTL, but that careful consideration of population size and experimental design are required.

Why Psychologists Should by Default Use Welch’s t-test Instead of Student’s t-test

“…Unequal sample sizes are particularly common when examining measured variables, where it is not always possible to determine a priori how many of the collected subjects will fall in each category (e.g., sex, nationality, or marital status). However, even with complete randomized assignment to conditions, where the same number of subjects are assigned to each condition, unequal sample sizes can emerge when participants have to be removed from the data analysis due to being outliers because the experimental protocol was not followed when collecting the data (Shaw & Mitchell-Olds, 1993) or due to missing values (Wang et al, 2012).…”

Mapping QTL for agronomic traits in breeding populations

“…For population mapping, Wang et al (2011) have recently compared balanced with unbalanced data sets and found that balanced data sets may be advantageous in reducing the number of false-positive QTL. Despite the potential problem of an inflated falsepositive rate, these results demonstrate that phenotypic data from breeding programs can be used for QTL detection without the need for specialized balanced experimental designs.…”

“…The application of the K matrix by mixed models has recently been shown to be sufficient for the analysis of breeding populations Wang et al 2011;Würschum et al 2011a, b). However, as the appropriate correction depends on the extent of the genotype-phenotype covariance this must be determined separately for each data set.…”

Identification of significant single nucleotide polymorphisms for resistance to maize rough dwarf disease in elite maize (Zea mays L.) inbred lines

“…So far, QTL studies for MRDD resistance in maize have been mainly based on linkage mapping; most QTLs are population-specific, and the genetic variation detected in a specific bi-parental population may not be transferable to other populations. By contrast, QTLs identified in elite breeding germplasms are of direct relevance for crop improvement via knowledge-based breeding, and can be immediately used for MAS approaches (Würschum 2012;Wang et al 2012). In the present study, a set of 184 elite maize inbred lines from modern breeding programs was evaluated to identify QTLs associated with MRDD resistance in different years through genome-wide association analysis, with the aim of resistance-pyramiding to improve breeding efficiency by MAS.…”

“…A major goal of QTL mapping is to identify markertrait associations that can be used for MAS within breeding programs (Xu and Crouch 2008;Wang et al 2012). So far, QTL studies for MRDD resistance in maize have been mainly based on linkage mapping; most QTLs are population-specific, and the genetic variation detected in a specific bi-parental population may not be transferable to other populations.…”