BLUP
BLUP genetic evaluation 2020
An important goal of the Beenomix 2.0 project has already been reached in March 2020, despite the ongoing pandemic, generating the genetic indexes on the basis of the phenotypes collected in the previous period of time (2019). Other than the phenotypic data, every family relation acquired through the previous Beenomix project have been included in the evaluation. The record file had this structure:
| Types of bees | Son | Mother | Father |
|---|---|---|---|
| DPQ | 2 | yes | no |
| DPQ | 4 | yes | yes |
| Queens | 18 | no | yes |
| Queens | 240 | yes | no |
| Queens | 276 | yes | yes |
| Phenotypic Colonies | 516 | yes | yes |
| Total | 1056 | 1038 | 796 |
A total of 1056 individuals with a mother and 796 with a father. Data distribution per year was the following. It shoul be noted that the selective scheme at the beginning was on 108 colonies.
| Year | Phenotypic Colonies |
| 2015 | 132 |
| 2016 | 72 |
| 2017 | 105 |
| 2018 | 107 |
| 2019 | 100 |
| Total | 516 |
According to the kinship matrix, the work group (Stefano Biffani, Giulio Pagnacco and Maria Grazia De Iorio) calculated the BLUP-ST indexes for all the 2015 archive onwards. Overall, there were 394 data per kg of honey, 246 for docility and 235 for hygienic behaviour. The genetic indices are defined as Estimated Breeding Value (EBV)Estimated Breeding Value).
The model, the classic German one, includes: general average, environmental effects of the year, of the apiary and of the interaction between these two. Secondly, the genetic effects of the worker bees of a certain family (direct effect on production) and of the queen who leads that family (maternal effect), in addition to the error effect. The index of a certain colony is given by the sum between the direct and the maternal effect of that colony.
The indices for the three characters were then weighted into a single aggregate index called IMEL, in which the weights attributed to the three characters reflect the following selective emphases:
1 for the kg of honey
0.8 for the docility
0.4 for the hygienic behaviour
Since the aggregate index does not have a defined unit of measurement, it was rescaled using the variability of the kilograms of honey, which is certainly more familiar with beekeepers. Below there are the correlations between indices and between these and the original phenotypes.
| Honey EBV | HB EBV | EBV docility | IMEL | Honey Kg | HB | docility | |
| Honey EBV | 1,00 | 0,24 | 0,09 | 0,68 | 0,93 | 0,23 | 0,09 |
| HB EBV | 1,00 | 0,22 | 0,38 | 0,26 | 0,99 | 0,16 | |
| EBV docility | 1,00 | 0,79 | 0,11 | 0,15 | 0,99 | ||
| IMEL | 1,00 | 0,66 | 0,32 | 0,77 | |||
| Honey Kg | 1,00 | 0,28 | 0,15 | ||||
| HB | 1,00 | 0,13 | |||||
| docility | 1,00 |
The correlations are all positive (a very advantageous fact): therefore, by selecting one of the characters the others follow it favourably. Corresponding indices and phenotypes are very well correlated as expected. The correlations between the indices are not true genetic correlations, but they approximate them. From the ranking of colonies measured in 2019, those from which the necessary translocations should have been made (2020) to constitute the new selection population were identified. In particular, the maternal grandmother from which to raise the future DPQs and the mothers of the 108 queens whose colonies will be evaluated in 2021 was identified.
The genetic evaluations were carried out by obtaining beforehand the estimates of the variances of the random components of the model (VQ, VW and VE), as well as the covariance between Q and W (Cov Q, W).Q, VW e VE), as well as the covariance between Q and W (CovQ,W).
