Comparing performances of different statistical models and multiple threshold methods in a nested association mapping...

Area of Science:

• Plant genetics and breeding
• Statistical genomics
• Quantitative trait analysis

Background:

• Nested association mapping (NAM) populations integrate biparental linkage mapping and association mapping for enhanced allelic richness and statistical power.
• Various statistical models and significance threshold methods exist for genome-wide association studies (GWAS), but their performance in NAM populations varies.
• Accurate detection of marker-trait associations (MTAs) requires robust statistical approaches to control false positives and negatives.

Purpose of the Study:

• To compare the performance of seven statistical models for MTAs in a spring wheat (Triticum aestivum L.) NAM population.
• To identify the optimal statistical model and significance threshold method for association analysis in NAM populations.
• To detect MTAs for agronomic and spectral reflectance traits using the best-performing methods.

Main Methods:

• Evaluated seven statistical models, including single and multi-locus approaches, using eight simulated traits with diverse genetic architectures in a wheat NAM population.
• Assessed model performance using QQ plots and the detection of true associations, false positives, and false negatives.
• Compared multiple significance threshold methods, including Bonferroni correction, for controlling spurious associations.

Main Results:

• The Bayesian information and linkage disequilibrium iteratively nested keyway (BLINK) model demonstrated superior performance over other models in simulated data, effectively controlling both false positives and false negatives.
• Bonferroni correction outperformed other threshold methods in controlling false positives and false negatives, complementing the effectiveness of GWAS models.
• BLINK, in conjunction with Bonferroni correction (0.05 significance threshold), identified 45 MTAs for 11 phenotypic traits in the wheat NAM population.

Conclusions:

• The BLINK model is a highly effective statistical tool for association analysis in NAM populations, offering robust control over statistical errors.
• Bonferroni correction is a reliable significance threshold method for enhancing the accuracy of GWAS in NAM studies.
• This study provides a methodological framework for future association analyses in NAM populations, aiding in the identification of genetic loci underlying important traits.