How partial phenotyping to reduce generation intervals can help to increase annual genetic gain in selected honeybee populations

Abstract

Honeybee breeding is organized around annual cycles, following seasonal change. Generation intervals are thus commonly multiples of whole years. Most queens are generally raised during spring or early summer in temperate climates. A generation interval of 1 year limits phenotyping to early recordable traits, before the spring following queens births. Some traits, however, can only be recorded later, as is typically the case for total honey yield. Their recording on selection candidates thus increases the generation interval to at least 2 years, a common interval on the dam path. Using stochastic simulation, we investigated the impact of halving the dam generation interval and therefore recording only early traits on candidate dams. The generation interval on the sire path remained at 2 years with complete phenotyping. Breeding goals with varying weights on early and late traits were considered, as well as different genetic correlations between traits, from negative to positive. The acceleration of the breeding scheme generally resulted in an increased genetic gain for the two-trait breeding goals, from 0% up to +47% after 20 years of selection. Although inbreeding rates per generation were slightly lower in the accelerated breeding scheme, associate inbreeding levels were about 20-30% higher after 20 years of selection due to the 33% increased generation turnover. To avoid too high inbreeding, shortening the generation interval should thus be accompanied by strategies to limit inbreeding while still retaining most of the genetic gain, such as increasing the breeding nucleus size by relaxing selection intensity.