Uncertainty in the mating strategy of honeybees causes bias and unreliability in the estimates of genetic parameters.

BACKGROUND: Breeding queens may be mated with drones that are produced by a single drone-producing queen (DPQ), or a group of sister-DPQs, but often only the dam of the DPQ(s) is reported in the pedigree. Furthermore, datasets may include colony phenotypes from DPQs that were open-mated at different locations, and thus to a heterogeneous drone population. METHODS: Simulation was used to investigate the impact of the mating strategy and its modelling on the estimates of genetic parameters and genetic trends when the DPQs are treated in different ways in the statistical evaluation model. We quantified the bias and standard error of the estimates when breeding queens were mated to one DPQ or a group of DPQs, assuming that this information was known or not. We also investigated four alternative strategies to accommodate the phenotypes of open-mated DPQs in the genetic evaluation: excluding their phenotypes, adding a dummy pseudo-sire in the pedigree, or adding a non-genetic (fixed or random) effect to the statistical evaluation model to account for the origin of the mates. RESULTS: The most precise estimates of genetic parameters and genetic trends were obtained when breeding queens were mated with drones of single DPQs that are correctly assigned in the pedigree. However, when they were mated with drones from one or a group of DPQs, and this information was not known, erroneous assumptions led to considerable bias in these estimates. Furthermore, genetic variances were considerably overestimated when phenotypes of colonies from open-mated DPQs were adjusted for their mates by adding a dummy pseudo-sire in the pedigree for each subpopulation of open-mating drones. On the contrary, correcting for the heterogeneous drone population by adding a non-genetic effect in the evaluation model produced unbiased estimates. CONCLUSIONS: Knowing only the dam of the DPQ(s) used in each mating may lead to erroneous assumptions on how DPQs were used and severely bias the estimates of genetic parameters and trends. Thus, we recommend keeping track of DPQs in the pedigree, and not only of the dams of DPQ(s). Records from DPQ colonies with queens open-mated to a heterogeneous drone population can be integrated by adding non-genetic effects to the statistical evaluation model.

Exploring Two Honey Bee Traits for Improving Resistance Against Varroa destructor: Development and Genetic Evaluation.

For the development of novel selection traits in honey bees, applicability under field conditions is crucial. We thus evaluated two novel traits intended to provide resistance against the ectoparasitic mite Varroa destructor and to allow for their straightforward implementation in honey bee selection. These traits are new field estimates of already-described colony traits: brood recapping rate ('Recapping') and solidness ('Solidness'). 'Recapping' refers to a specific worker characteristic wherein they reseal a capped and partly opened cell containing a pupa, whilst 'Solidness' assesses the percentage of capped brood in a predefined area. According to the literature and beekeepers' experiences, a higher recapping rate and higher solidness could be related to resistance to V. destructor. During a four-year field trial in Switzerland, the two resistance traits were assessed in a total of 121 colonies of Apis mellifera mellifera. We estimated the repeatability and the heritability of the two traits and determined their phenotypic correlations with commonly applied selection traits, including other putative resistance traits. Both traits showed low repeatability between different measurements within each year. 'Recapping' had a low heritability (h2 = 0.04 to 0.05, depending on the selected model) and a negative phenotypic correlation to non-removal of pin-killed brood (r = -0.23). The heritability of 'Solidness' was moderate (h2 = 0.24 to 0.25) and did not significantly correlate with resistance traits. The two traits did not show an association with V. destructor infestation levels. Further research is needed to confirm the results, as only a small number of colonies was evaluated.

Heritability estimates of the novel trait 'suppressed in ovo virus infection' in honey bees (Apis mellifera).

Honey bees are under pressure due to abnormal high colony death rates, especially during the winter. The infestation by the Varroa destructor mite and the viruses that this ectoparasite transmits are generally considered as the bees' most important biological threats. Almost all efforts to remedy this dual infection have so far focused on the control of the Varroa mite alone and not on the viruses it transmits. In the present study, the sanitary control of breeding queens was conducted on eggs taken from drone brood for 4 consecutive years (2015-2018). The screening was performed on the sideline of an ongoing breeding program, which allowed us to estimate the heritabilities of the virus status of the eggs. We used the term 'suppressed in ovo virus infection' (SOV) for this novel trait and found moderate heritabilities for the presence of several viruses simultaneously and for the presence of single viral species. Colonies that expressed the SOV trait seemed to be more resilient to virus infections as a whole with fewer and less severe Deformed wing virus infections in most developmental stages, especially in the male caste. The implementation of this novel trait into breeding programs is recommended.

Evaluation of Traits for the Selection of Apis Mellifera for Resistance against Varroa Destructor.

Infestation with Varroa destructor is a serious cause of bee colony (Apis mellifera) losses on a global level. However, the presence of untreated survivor populations in many different regions supports the idea that selection for resistance can be successful. As colony survival is difficult or impossible to measure, differences in mite infestation levels and tests for specific behavioral traits are used for selective breeding for Varroa resistance. In this paper we looked into different definitions of mite infestation and linked these with brood hygiene (pin test), brood recapping and suppressed mite reproduction. We based our analyses on datasets of Apis mellifera carnica from three countries: Austria (147 records), Croatia (135) and Germany (207). We concluded that bee infestation in summer, adjusted for the level of natural mite fall in spring, is a suitable trait in the breeding objective, and also suggested including brood infestation rate and the increase rate of bee infestation in summer. Repeatability for bee infestation rate was about 0.55, for cells opened in pin test about 0.33, for recapping 0.35 and for suppressed mite reproduction (SMR) virtually zero. Although in most cases we observed correlations with the expected sign between infestation parameters and behavioral traits, the values were generally low (<0.2) and often not significantly different from zero.

A note on genetic parameters and accuracy of estimated breeding values in honey bees.

BACKGROUND: In honey bees, observations are usually made on colonies. The phenotype of a colony is affected by the average breeding value for the worker effect of the thousands of workers in the colony (the worker group) and by the breeding value for the queen effect of the queen of the colony. Because the worker group consists of multiple individuals, interpretation of the variance components and heritabilities of phenotypes observed on the colony and of the accuracy of selection is not straightforward. The additive genetic variance among worker groups depends on the additive genetic relationship between the drone-producing queens (DPQ) that produce the drones that mate with the queen. RESULTS: Here, we clarify how the relatedness between DPQ affects phenotypic variance, heritability and accuracy of the estimated breeding values of replacement queens. Second, we use simulation to investigate the effect of assumptions about the relatedness between DPQ in the base population on estimates of genetic parameters. Relatedness between DPQ in the base generation may differ considerably between populations because of their history. CONCLUSIONS: Our results show that estimates of (co)variance components and derived genetic parameters were seriously biased (25% too high or too low) when assumptions on the relationship between DPQ in the statistical analysis did not agree with reality.

Methods to estimate breeding values in honey bees.

BACKGROUND: Efficient methodologies based on animal models are widely used to estimate breeding values in farm animals. These methods are not applicable in honey bees because of their mode of reproduction. Observations are recorded on colonies, which consist of a single queen and thousands of workers that descended from the queen mated to 10 to 20 drones. Drones are haploid and sperms are copies of a drone's genotype. As a consequence, Mendelian sampling terms of full-sibs are correlated, such that the covariance matrix of Mendelian sampling terms is not diagonal. RESULTS: In this paper, we show how the numerator relationship matrix and its inverse can be obtained for honey bee populations. We present algorithms to derive the covariance matrix of Mendelian sampling terms that accounts for correlated terms. The resulting matrix is a block-diagonal matrix, with a small block for each full-sib family, and is easy to invert numerically. The method allows incorporating the within-colony distribution of progeny from drone-producing queens and drones, such that estimates of breeding values weigh information from relatives appropriately. Simulation shows that the resulting estimated breeding values are unbiased predictors of true breeding values. Benefits for response to selection, compared to an existing approximate method, appear to be limited (~5%). Benefits may however be greater when estimating genetic parameters. CONCLUSIONS: This work shows how the relationship matrix and its inverse can be developed for honey bee populations, and used to estimate breeding values and variance components.