# Population-wide modelling reveals prospects of marker-assisted selection for parasitic mite resistance in honey bees

## TL;DR

Researchers tested if genetic markers for mite resistance in honey bees work across a broader population.

## Contribution

A reduced three-variant model with 76% classification accuracy was developed for mite resistance in honey bees.

## Key findings

- A three-variant model correctly classified 76% of Varroa-infested drone brood resistance cases.
- Allelic frequencies of three variants were determined in 292 A. mellifera carnica queens.
- The study shows potential for marker-assisted selection of mite-resistant honey bees in Flanders.

## Abstract

In 2019, a joint eight-variant model was published in which eight single nucleotide polymorphisms (SNPs) in seven Apis mellifera genes were associated with Varroa destructor drone brood resistance (DBR, i.e. mite non-reproduction in drone brood). As this model was derived from only one Darwinian Black Bee Box colony, it could not directly be applied on a population-overarching scale in the northern part of Belgium (Flanders), where beekeepers prefer the carnica subspecies. To determine whether these eight SNPs remained associated with the DBR trait on a Flemish colony-broad scope, we performed population-wide modelling through sampling of various A. mellifera carnica colonies, DBR scoring of Varroa-infested drone brood and variant genotyping. Novel eight-variant modelling was performed and the classification performance of the eight SNPs was evaluated. Besides, we built a reduced three-variant model retaining only three genetic variants and found that this model classified 76% of the phenotyped drones correctly. To examine the spread of beneficial alleles and predict the DBR probability distribution in Flanders, we determined the allelic frequencies of the three variants in 292 A. mellifera carnica queens. As such, this research reveals prospects of marker-assisted selection for Varroa drone brood resistance in honeybees.