Bayesian Tests for Segregation Distortion in Experimental Tetraploid Populations

Here, we focus on testing for segregation distortion in F1 populations of tetraploids. Tetraploids, which contain four sets of chromosomes, are important in both evolutionary research and agricultural improvement. In agricultural experiments with F1 populations, as a quality control measure, researchers often test whether genotype frequencies conform to those expected under Mendelian segregation. However, classical segregation patterns can be distorted in tetraploids due to the meiotic processes of double reduction and preferential pairing, which can alter gamete frequencies. Currently, there is no method to test for segregation distortion while accounting for these two processes in tetraploid F1 populations, despite their widespread use in agriculture. To address this gap, we propose a Bayesian approach that incorporates both preferential pairing and double reduction in a new model for offspring genotypes in tetraploid F1 populations. We demonstrate the efficacy of our approach through simulations and a real dataset of tetraploid blueberries. Our method inherits all of the benefits of Bayesian analysis, including consistency under the null, incorporation of prior information, and applicability to small samples by not depending on asymptotic approximations. This study provides a valuable contribution to the field by offering a new tool for testing for segregation distortion in tetraploid F1 populations.