Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches
Abstract Background Genetic improvement of honey bees is more difficult compared to other livestock, due to the very different reproductive behavior. Estimation of breeding values requires specific adjustment and the use of sires in the pedigree is only possible when mating of queens and drones is s...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2019-10-01
|
| Series: | BMC Genetics |
| Subjects: | |
| Online Access: | http://link.springer.com/article/10.1186/s12863-019-0776-2 |
| _version_ | 1811337992842772480 |
|---|---|
| author | Sreten Andonov Cecilia Costa Aleksandar Uzunov Patrizia Bergomi Daniela Lourenco Ignacy Misztal |
| author_facet | Sreten Andonov Cecilia Costa Aleksandar Uzunov Patrizia Bergomi Daniela Lourenco Ignacy Misztal |
| author_sort | Sreten Andonov |
| collection | DOAJ |
| description | Abstract Background Genetic improvement of honey bees is more difficult compared to other livestock, due to the very different reproductive behavior. Estimation of breeding values requires specific adjustment and the use of sires in the pedigree is only possible when mating of queens and drones is strictly controlled. In the breeding program of the National Registry for Italian Queen Breeders and Bee Producers the paternal contribution is mostly unknown. As stronger modeling may compensate for the lack of pedigree information, we tested two models that differed in the way the direct and maternal effects were considered. The two models were tested using 4003 records for honey yield, defensive and swarming behaviors of Italian honey bee queens produced between 2002 and 2014. The first model accounted for the direct genetic effect of worker bees and the genetic maternal effect of the queen, whereas model 2 considered the direct genetic effect of the queen without maternal effect. The analyses were performed by linear (honey production) and threshold (defensive and swarming behavior) single-trait models; estimated genetic correlations among traits were obtained by a three-trait linear-threshold model. Results For all traits, the highest predictability (correlation between breeding values estimated with and without performance records) was obtained with model 2, where direct genetic effect of queens was considered. With this model, heritability estimates were 0.26 for honey yield, 0.36 for defensive behavior, and 0.34 for swarming behavior. Multi-trait estimation resulted in similar or higher heritability estimates for all traits. A low, positive genetic correlation (0.19) was found between honey yield and defensive behavior, whereas the genetic correlation between honey yield and swarming behavior was moderate (0.41). A strong, positive genetic correlation was found between defensive and swarming behaviors (0.62). Predictability for multi-trait evaluations was higher for honey yield (0.46) and defensive behavior (0.30) but almost identical for swarming behavior (0.45) compared to corresponding single-trait predictability. Conclusions Multi-trait evaluation using a model that accounts for the direct genetic effect of queen was the best approach for breeding value estimation of Italian honey bees. The results suggest a new direction for selection of linear and categorical traits in breeding programs where drone origin is unknown. |
| first_indexed | 2024-04-13T18:04:12Z |
| format | Article |
| id | doaj.art-075970e196f54eedbb276c8838689def |
| institution | Directory Open Access Journal |
| issn | 1471-2156 |
| language | English |
| last_indexed | 2024-04-13T18:04:12Z |
| publishDate | 2019-10-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Genetics |
| spelling | doaj.art-075970e196f54eedbb276c8838689def2022-12-22T02:36:07ZengBMCBMC Genetics1471-21562019-10-012011910.1186/s12863-019-0776-2Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approachesSreten Andonov0Cecilia Costa1Aleksandar Uzunov2Patrizia Bergomi3Daniela Lourenco4Ignacy Misztal5Department of Animal Breding and Genetic, Swedish University of Animal SceincesCREA Research Centre for Agriculture and EnvironmentFaculty of Agricultural Sciences and Food, Ss. Cyril and Methodius UniversityCREA Research Centre for Agriculture and EnvironmentDepartment of Animal and Dairy Science, University of GeorgiaDepartment of Animal and Dairy Science, University of GeorgiaAbstract Background Genetic improvement of honey bees is more difficult compared to other livestock, due to the very different reproductive behavior. Estimation of breeding values requires specific adjustment and the use of sires in the pedigree is only possible when mating of queens and drones is strictly controlled. In the breeding program of the National Registry for Italian Queen Breeders and Bee Producers the paternal contribution is mostly unknown. As stronger modeling may compensate for the lack of pedigree information, we tested two models that differed in the way the direct and maternal effects were considered. The two models were tested using 4003 records for honey yield, defensive and swarming behaviors of Italian honey bee queens produced between 2002 and 2014. The first model accounted for the direct genetic effect of worker bees and the genetic maternal effect of the queen, whereas model 2 considered the direct genetic effect of the queen without maternal effect. The analyses were performed by linear (honey production) and threshold (defensive and swarming behavior) single-trait models; estimated genetic correlations among traits were obtained by a three-trait linear-threshold model. Results For all traits, the highest predictability (correlation between breeding values estimated with and without performance records) was obtained with model 2, where direct genetic effect of queens was considered. With this model, heritability estimates were 0.26 for honey yield, 0.36 for defensive behavior, and 0.34 for swarming behavior. Multi-trait estimation resulted in similar or higher heritability estimates for all traits. A low, positive genetic correlation (0.19) was found between honey yield and defensive behavior, whereas the genetic correlation between honey yield and swarming behavior was moderate (0.41). A strong, positive genetic correlation was found between defensive and swarming behaviors (0.62). Predictability for multi-trait evaluations was higher for honey yield (0.46) and defensive behavior (0.30) but almost identical for swarming behavior (0.45) compared to corresponding single-trait predictability. Conclusions Multi-trait evaluation using a model that accounts for the direct genetic effect of queen was the best approach for breeding value estimation of Italian honey bees. The results suggest a new direction for selection of linear and categorical traits in breeding programs where drone origin is unknown.http://link.springer.com/article/10.1186/s12863-019-0776-2Threshold modelHeritabilityGenetic correlationPredictabilityGenetic evaluation |
| spellingShingle | Sreten Andonov Cecilia Costa Aleksandar Uzunov Patrizia Bergomi Daniela Lourenco Ignacy Misztal Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches BMC Genetics Threshold model Heritability Genetic correlation Predictability Genetic evaluation |
| title | Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches |
| title_full | Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches |
| title_fullStr | Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches |
| title_full_unstemmed | Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches |
| title_short | Modeling honey yield, defensive and swarming behaviors of Italian honey bees (Apis mellifera ligustica) using linear-threshold approaches |
| title_sort | modeling honey yield defensive and swarming behaviors of italian honey bees apis mellifera ligustica using linear threshold approaches |
| topic | Threshold model Heritability Genetic correlation Predictability Genetic evaluation |
| url | http://link.springer.com/article/10.1186/s12863-019-0776-2 |
| work_keys_str_mv | AT sretenandonov modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches AT ceciliacosta modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches AT aleksandaruzunov modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches AT patriziabergomi modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches AT danielalourenco modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches AT ignacymisztal modelinghoneyyielddefensiveandswarmingbehaviorsofitalianhoneybeesapismelliferaligusticausinglinearthresholdapproaches |