Spatial modelling improves genetic evaluation in smallholder breeding programs

Livestock phenomics and genetic evaluation approaches in Africa: current state and future perspectives

The African livestock sector plays a key role in improving the livelihoods of people through the supply of food, improved nutrition and consequently health. However, its impact on the economy of the people and contribution to national GDP is highly variable and generally below its potential. This study was conducted to assess the current state of livestock phenomics and genetic evaluation methods being used across the continent, the main challenges, and to demonstrate the effects of various genetic models on the accuracy and rate of genetic gain that could be achieved. An online survey of livestock experts, academics, scientists, national focal points for animal genetic resources, policymakers, extension agents and animal breeding industry was conducted in 38 African countries. The results revealed 1) limited national livestock identification and data recording systems, 2) limited data on livestock production and health traits and genomic information, 3) mass selection was the common method used for genetic improvement with very limited application of genetic and genomic-based selection and evaluation, 4) limited human capacity, infrastructure, and funding for livestock genetic improvement programmes, as well as enabling animal breeding policies. A joint genetic evaluation of Holstein-Friesian using pooled data from Kenya and South Africa was piloted. The pilot analysis yielded higher accuracy of prediction of breeding values, pointing to possibility of higher genetic gains that could be achieved and demonstrating the potential power of multi-country evaluations: Kenya benefited on the 305-days milk yield and the age at first calving and South Africa on the age at first calving and the first calving interval. The findings from this study will help in developing harmonized protocols for animal identification, livestock data recording, and genetic evaluations (both national and across-countries) as well as in designing subsequent capacity building and training programmes for animal breeders and livestock farmers in Africa. National governments need to put in place enabling policies, the necessary infrastructure and funding for national and across country collaborations for a joint genetic evaluation which will revolutionize the livestock genetic improvement in Africa.

SIMplyBee: an R package to simulate honeybee populations and breeding programs

BACKGROUND

The Western honeybee is an economically important species globally, but has been experiencing colony losses that lead to economical damage and decreased genetic variability. This situation is spurring additional interest in honeybee breeding and conservation programs. Stochastic simulators are essential tools for rapid and low-cost testing of breeding programs and methods, yet no existing simulator allows for a detailed simulation of honeybee populations. Here we describe SIMplyBee, a holistic simulator of honeybee populations and breeding programs. SIMplyBee is an R package and hence freely available for installation from CRAN http://cran.r-project.org/package=SIMplyBee .

IMPLEMENTATION

SIMplyBee builds upon the stochastic simulator AlphaSimR that simulates individuals with their corresponding genomes and quantitative genetic values. To enable honeybee-specific simulations, we extended AlphaSimR by developing classes for global simulation parameters, SimParamBee, for a honeybee colony, Colony, and multiple colonies, MultiColony. We also developed functions to address major honeybee specificities: honeybee genome, haplodiploid inheritance, social organisation, complementary sex determination, polyandry, colony events, and quantitative genetics at the individual- and colony-levels.

RESULTS

We describe its implementation for simulating a honeybee genome, creating a honeybee colony and its members, addressing haplodiploid inheritance and complementary sex determination, simulating colony events, creating and managing multiple colonies at the same time, and obtaining genomic data and honeybee quantitative genetics. Further documentation, available at http://www.SIMplyBee.info , provides details on these operations and describes additional operations related to genomics, quantitative genetics, and other functionalities.

DISCUSSION

SIMplyBee is a holistic simulator of honeybee populations and breeding programs. It simulates individual honeybees with their genomes, colonies with colony events, and individual- and colony-level genetic and breeding values. Regarding the latter, SIMplyBee takes a user-defined function to combine individual- into colony-level values and hence allows for modeling any type of interaction within a colony. SIMplyBee provides a research platform for testing breeding and conservation strategies and their effect on future genetic gain and genetic variability. Future developments of SIMplyBee will focus on improving the simulation of honeybee genomes, optimizing the simulator's performance, and including spatial awareness in mating functions and phenotype simulation. We invite the honeybee genetics and breeding community to join us in the future development of SIMplyBee.

Use of Milk Infrared Spectral Data as Environmental Covariates in Genomic Prediction Models for Production Traits in Canadian Holstein

The purpose of this study was to provide a procedure for the inclusion of milk spectral information into genomic prediction models. Spectral data were considered a set of covariates, in addition to genomic covariates. Milk yield and somatic cell score were used as traits to investigate. A cross-validation was employed, making a distinction for predicting new individuals' performance under known environments, known individuals' performance under new environments, and new individuals' performance under new environments. We found an advantage of including spectral data as environmental covariates when the genomic predictions had to be extrapolated to new environments. This was valid for both observed and, even more, unobserved families (genotypes). Overall, prediction accuracy was larger for milk yield than somatic cell score. Fourier-transformed infrared spectral data can be used as a source of information for the calculation of the 'environmental coordinates' of a given farm in a given time, extrapolating predictions to new environments. This procedure could serve as an example of integration of genomic and phenomic data. This could help using spectral data for traits that present poor predictability at the phenotypic level, such as disease incidence and behavior traits. The strength of the model is the ability to couple genomic with high-throughput phenomic information.

Genomic evaluation of milk yield in a smallholder crossbred dairy production system in India

BACKGROUND

India is the largest milk producer globally, with the largest proportion of cattle milk production coming from smallholder farms with an average herd size of less than two milking cows. These cows are mainly undefined multi-generation crosses between exotic dairy breeds and indigenous Indian cattle, with no performance or pedigree recording. Therefore, implementing genetic improvement based on genetic evaluation has not yet been possible. We present the first results from a large smallholder performance recording program in India, using single nucleotide polymorphism (SNP) genotypes to estimate genetic parameters for monthly test-day (TD) milk records and to obtain and validate genomic estimated breeding values (GEBV).

RESULTS

The average TD milk yield under the high, medium, and low production environments were 9.64, 6.88, and 4.61 kg, respectively. In the high production environment, the usual profile of a lactation curve was evident, whereas it was less evident in low and medium production environments. There was a clear trend of an increasing milk yield with an increasing Holstein Friesian (HF) proportion in the high production environment, but no increase above intermediate grades in the medium and low production environments. Trends for Jersey were small but yield estimates had a higher standard error than HF. Heritability estimates for TD yield across the lactation ranged from 0.193 to 0.250, with an average of 0.230. The additive genetic correlations between TD yield at different times in lactation were high, ranging from 0.846 to 0.998. The accuracy of phenotypic validation of GEBV from the method that is believed to be the least biased was 0.420, which was very similar to the accuracy obtained from the average prediction error variance of the GEBV.

CONCLUSIONS

The results indicate strong potential for genomic selection to improve milk production of smallholder crossbred cows in India. The performance of cows with different breed compositions can be determined in different Indian environments, which makes it possible to provide better advice to smallholder farmers on optimum breed composition for their environment.

GPS Coordinates for Modelling Correlated Herd Effects in Genomic Prediction Models Applied to Hanwoo Beef Cattle

Simple Summary

It is widely known that the environment influences phenotypic expression and that its effects must be accounted for in genetic evaluation programs. The most used method to account for environmental effects is to add herd and the contemporary group to the model. Although generally informative, the herd effect treats different farms as independent units. However, if two farms are located physically close to each other, they potentially share correlated environmental factors. We introduce a method to model herd effects using physical distances between farms based on GPS coordinates as a proxy for the correlation matrix of these effects, aiming to account for similarities and differences between farms due to environmental factors. A population of beef cattle was used to evaluate the impact on the variance components and on the genomic prediction, of modelling herd effects as correlated, in comparison to assuming the farms as completely independent units. The main result was an increase in the reliabilities of the predicted genomic breeding values compared to reliabilities obtained with traditional models, a finding of practical relevance for genetic evaluation programs.

Abstract

It is widely known that the environment influences phenotypic expression and that its effects must be accounted for in genetic evaluation programs. The most used method to account for environmental effects is to add herd and contemporary group to the model. Although generally informative, the herd effect treats different farms as independent units. However, if two farms are located physically close to each other, they potentially share correlated environmental factors. We introduce a method to model herd effects that uses the physical distances between farms based on the Global Positioning System (GPS) coordinates as a proxy for the correlation matrix of these effects that aims to account for similarities and differences between farms due to environmental factors. A population of Hanwoo Korean cattle was used to evaluate the impact of modelling herd effects as correlated, in comparison to assuming the farms as completely independent units, on the variance components and genomic prediction. The main result was an increase in the reliabilities of the predicted genomic breeding values compared to reliabilities obtained with traditional models (across four traits evaluated, reliabilities of prediction presented increases that ranged from 0.05 ± 0.01 to 0.33 ± 0.03), suggesting that these models may overestimate heritabilities. Although little to no significant gain was obtained in phenotypic prediction, the increased reliability of the predicted genomic breeding values is of practical relevance for genetic evaluation programs.