Evaluation of nine statistics to identify QTLs in bulk segregant analysis using next generation sequencing approaches

Abstract

Abstract Background Bulk segregant analysis (BSA) combined with next generation sequencing is a powerful tool to identify quantitative trait loci (QTL). The impact of the size of the study population and the percentage of extreme genotypes analysed have already been assessed. But a good comparison of statistical approaches designed to identify QTL regions using next generation sequencing (NGS) technologies for BSA is still lacking. Results We developed an R code to simulate QTLs in bulks of F2 contrasted lines. We simulated a range of recombination rates based on estimations using different crop species. The simulations were used to benchmark the ability of statistical methods identify the exact location of true QTLs. A single QTL led to a shift in allele frequency across a large fraction of the chromosome for plant species with low recombination rate. The smoothed version of all statistics performed best notably the smoothed Euclidean distance-based statistics was always found to be more accurate in identifying the location of QTLs. We propose a simulation approach to build confidence interval statistics for the detection of QTLs. Conclusion We highlight the statistical methods best suited for BSA studies using NGS technologies in crops even when recombination rate is low. We also provide simulation codes to build confidence intervals and to assess the impact of recombination for application to other studies. This computational study will help select NGS-based BSA statistics that are useful to the broad scientific community.