The feasibility of using low-density marker panels for genotype imputation and genomic prediction of crossbred dairy cattle of East Africa.

Cost-effective high-density (HD) genotypes of livestock species can be obtained by genotyping a proportion of the population using a HD panel and the remainder using a cheaper low-density panel, and then imputing the missing genotypes that are not directly assayed in the low-density panel. The efficacy of genotype imputation can largely be affected by the structure and history of the specific target population and it should be checked before incorporating imputation in routine genotyping practices. Here, we investigated the efficacy of imputation in crossbred dairy cattle populations of East Africa using 4 different commercial single nucleotide polymorphisms (SNP) panels, 3 reference populations, and 3 imputation algorithms. We found that Minimac and a reference population, which included a mixture of crossbred and ancestral purebred animals, provided the highest imputation accuracy compared with other scenarios of imputation. The accuracies of imputation, measured as the correlation between real and imputed genotypes averaged across SNP, were around 0.76 and 0.94 for 7K and 40K SNP, respectively, when imputed up to a 770K panel. We also presented a method to maximize the imputation accuracy of low-density panels, which relies on the pairwise (co)variances between SNP and the minor allele frequency of SNP. The performance of the developed method was tested in a 5-fold cross-validation process where various densities of SNP were selected using the (co)variance method and also by alternative SNP selection methods and then imputed up to the HD panel. The (co)variance method provided the highest imputation accuracies at almost all marker densities, with accuracies being up to 0.19 higher than the random selection of SNP. The accuracies of imputation from 7K and 40K panels selected using the (co)variance method were around 0.80 and 0.94, respectively. The presented method also achieved higher accuracy of genomic prediction at lower densities of selected SNP. The squared correlation between genomic breeding values estimated using imputed genotypes and those from the real 770K HD panel was 0.95 when the accuracy of imputation was 0.64. The presented method for SNP selection is straightforward in its application and can ensure high accuracies in genotype imputation of crossbred dairy populations in East Africa.

Purebreeding of Red Maasai and crossbreeding with Dorper sheep in different environments in Kenya.

The aim of this article was to study opportunities for improvement of the indigenous and threatened Red Maasai sheep (RM) in Kenya, by comparing purebreeding with crossbreeding with Dorper sheep (D) as a terminal breed, in two different environments (Env. A and a harsher Env. B), assuming different levels of genotype-by-environment interaction (G × E). Breeding goals differed between environments and breeds. Four scenarios of nucleus breeding schemes were stochastically simulated, with the nucleus in Env. A. Overall, results showed an increase in carcass weight produced per ewe by more than 10% over 15 years. Genetic gain in carcass weight was 0.17 genetic SD/year (0.2 kg/year) across scenarios for RM in the less harsh Env. A. For survival and milk yield, the gain was lower (0.04-0.05 genetic SD/year). With stronger G × E, the gain in the commercial tier for RM in the harsher Env. B became increasingly lower. Selection of females also within the commercial tier gave slightly higher genetic gain. The scenario with purebreeding of RM and a subnucleus in Env. B gave the highest total income and quantity of meat. However, quantity of meat in Env. A increased slightly from having crossbreeding with D, whereas that in Env. B decreased. A simple and well-designed nucleus breeding programme would increase the genetic potential of RM. Crossbreeding of RM with D is not recommended for harsh environmental conditions due to the large breed differences expected in that environment.

Live weight, conformation, carcass traits and economic values of ram lambs of Red Maasai and Dorper sheep and their crosses.

Meat production is the most important trait in the breeding objectives of sheep production in East Africa. The aim of this study was to investigate breed differences in live weight, conformation, carcass traits and economic values for meat production among Red Maasai and Dorper sheep and their crosses. In total, 88 ram lambs, which were reared at the ILRI experimental station, Kapiti plains Estate in Central Kenya, were used for the study. The lambs were slaughtered at Kenya Meat Commission (KMC) at about 1 year of age. Prior to slaughter, the lambs were weighed, measured and assessed by experienced evaluators, and at the abattoir carcass traits were recorded. Large breed differences were found for most traits. Dorper lambs were heavier at delivery for slaughter and had better carcass grade but lower dressing percentage and fat levels than Red Maasai. Crossbreds were generally better than the parental breeds. Evaluators were willing to pay more for the Dorper lambs for slaughter although carcass weights later were shown not to be higher than for Red Maasai. Evaluators undervalued Red Maasai lambs by 8-13 % compared to Dorper lambs according to the prices quoted per kilogramme live or carcass weight by KMC. Live weight was better than any other live measure in predicting carcass weight. Due to the overall higher ranking of the crossbred lambs for meat production, Dorper may be useful as a terminal sire breed for crossing with Red Maasai ewes.

Genetic analysis of phenotypic indicators for heat tolerance in crossbred dairy cattle.

Climate change-induced rise in global temperatures has intensified heat stress on dairy cattle and is contributing to the generally observed low milk productivity. Selective breeding aimed at enhancing animals' ability to withstand rising temperatures while maintaining optimal performance is crucial for ensuring future access to dairy products. However, phenotypic indicators of heat tolerance are yet to be effectively factored into the objectives of most selective breeding programs. This study investigated the response of milk production to changing heat load as an indication of heat tolerance and the influence of calving season on this response in multibreed dairy cattle performing in three agroecological zones Kenya. First-parity 7-day average milk yield (65 261 milk records) of 1 739 cows were analyzed. Based on routinely recorded weather data that were accessible online, the Temperature-Humidity Index (THI) was calculated and used as a measure of heat load. THI measurements used represented averages of the same 7-day periods corresponding to each 7-day average milk record. Random regression models, including reaction norm functions, were fitted to derive two resilience indicators: slope of the reaction norm (Slope) and its absolute value (Absolute), reflecting changes in milk yield in response to the varying heat loads (THI 50 and THI 80). The genetic parameters of these indicators were estimated, and their associations with average test-day milk yield were examined. There were no substantial differences in the pattern of milk yield response to heat load between cows calving in dry and wet seasons. Animals with ≤50% Bos taurus genes were the most thermotolerant at extremely high heat load levels. Animals performing in semi-arid environments exhibited the highest heat tolerance capacity. Heritability estimates for these indicators ranged from 0.06 to 0.33 and were mostly significantly different from zero (P < 0.05). Slope at THI 80 had high (0.64-0.71) negative correlations with average daily milk yield, revealing that high-producing cows are more vulnerable to heat stress and vice versa. A high (0.63-0.74) positive correlation was observed between Absolute and average milk yield at THI 80. This implied that low milk-producing cows have a more stable milk production under heat-stress conditions and vice versa. The study demonstrated that the slope of the reaction norms and its absolute value can effectively measure the resilience of crossbred dairy cattle to varying heat load conditions. The implications of these findings are valuable in improving the heat tolerance of livestock species through genetic selection.

Accuracy of pastoralists' memory-based kinship assignment of Ankole cattle: a microsatellite DNA analysis.

This study aimed to estimate the level of relatedness within Ankole cattle herds using autosomal microsatellite markers and to assess the accuracy of relationship assignment based on farmers' memory. Eight cattle populations (four from each of two counties in Mbarara district in Uganda) were studied. Cattle in each population shared varying degrees of relatedness (first-, second- and third-degree relatives and unrelated individuals). Only memory-based kinship assignments which farmers knew with some confidence were tested in this experiment. DNA isolated from the blood of a subsample of 304 animals was analysed using 19 microsatellite markers. Average within population relatedness coefficients ranged from 0.010 ± 0.005 (Nshaara) to 0.067 ± 0.004 (Tayebwa). An exclusion probability of 99.9% was observed for both sire-offspring and dam-offspring relationships using the entire panel of 19 markers. Confidence from likelihood tests performed on 292 dyads showed that first-degree relatives were more easily correctly assigned by farmers than second-degree ones (p < 0.01), which were also easier to assign than third-degree relatives (p < 0.01). Accuracy of kinship assignment by the farmers was 91.9% ± 5.0 for dam-offspring dyads, 85.5% ± 3.4 for sire-offspring dyads, 75.6% ± 12.3 for half-sib and 60.0% ± 5.0 for grand dam-grand offspring dyads. Herd size, number of dyads assigned and length of time spent by the herder with their cattle population did not correlate with error in memorizing relationships. However, herd size strongly correlated with number of dyads assigned by the herder (r = 0.967, p < 0.001). Overall, we conclude that memorized records of pastoralists can be used to trace relationships and for pedigree reconstruction within Ankole cattle populations, but with the awareness that herd size constrains the number of kinship assignments remembered by the farmer.

Community-based alternative breeding plans for indigenous sheep breeds in four agro-ecological zones of Ethiopia.

Based on the results of participatory approaches to define traits in the breeding objectives, four scenarios of ram selection and ram use were compared via deterministic modelling of breeding plans for community-based sheep breeding programmes in four diverse agro-ecological regions of Ethiopia. The regions (and production systems) were Afar (pastoral/agro-pastoral), Bonga and Horro (both mixed crop-livestock) and Menz (sheep-barley). The schemes or scenarios differed in terms of selection intensity and duration of ram use. The predicted genetic gains per year in yearling weight (kilograms) were comparable across the schemes but differed among the breeds and ranged from 0.399 to 0.440 in Afar, 0.813 to 0.894 in Bonga, 0.850 to 0.940 in Horro, and 0.616 to 0.699 in Menz. The genetic gains per year in number of lambs born per ewe bred ranged from 0.009 to 0.010 in both Bonga and Horro. The predicted genetic gain in the proportion of lambs weaned per ewe joined was nearly comparable in all breeds ranging from 0.008 to 0.011. The genetic gain per year in milk yield of Afar breed was in the order of 0.018 to 0.020 kg, while the genetic gain per generation for greasy fleece weight (kg) ranged from 0.016 to 0.024 in Menz. Generally, strong selection and shorter duration of ram use for breeding were the preferred options. The expected genetic gains are satisfactory but largely rely on accurate and continuous pedigree and performance recording.

Ancestral Haplotype Mapping for GWAS and Detection of Signatures of Selection in Admixed Dairy Cattle of Kenya.

Understanding the genetic structure of adaptation and productivity in challenging environments is necessary for designing breeding programs that suit such conditions. Crossbred dairy cattle in East Africa resulting from over 60 years of crossing exotic dairy breeds with indigenous cattle plus inter se matings form a highly variable admixed population. This population has been subject to natural selection in response to environmental stresses, such as harsh climate, low-quality feeds, poor management, and strong disease challenge. Here, we combine two complementary sets of analyses, genome-wide association (GWA) and signatures of selection (SoS), to identify genomic regions that contribute to variation in milk yield and/or contribute to adaptation in admixed dairy cattle of Kenya. Our GWA separates SNP effects due to ancestral origin of alleles from effects due to within-population linkage disequilibrium. The results indicate that many genomic regions contributed to the high milk production potential of modern dairy breeds with no region having an exceptional effect. For SoS, we used two haplotype-based tests to compare haplotype length variation within admixed and between admixed and East African Shorthorn Zebu cattle populations. The integrated haplotype score (iHS) analysis identified 16 candidate regions for positive selection in the admixed cattle while the between population Rsb test detected 24 divergently selected regions in the admixed cattle compared to East African Shorthorn Zebu. We compare the results from GWA and SoS in an attempt to validate the most significant SoS results. Only four candidate regions for SoS intersect with GWA regions using a low stringency test. The identified SoS candidate regions harbored genes in several enriched annotation clusters and overlapped with previously found QTLs and associations for different traits in cattle. If validated, the GWA and SoS results indicate potential for SNP-based genomic selection for genetic improvement of smallholder crossbred cattle.

Evaluating the impact of heat stress as measured by temperature-humidity index (THI) on test-day milk yield of small holder dairy cattle in a sub-Sahara African climate.

This study evaluates the effect of heat stress on milk production and describes the pattern of response of milk yield to increasing heat load, using temperature-humidity index (THI) on test-day milk records of small holder dairy cattle herds in the sub-Saharan African climate of Tanzania. Climate data obtained from aWhere, an agricultural weather data platform (http://www.awhere.com) was analysed with 14,367 first lactation test day milk records of 3511 dairy cows collected between 2016 and 2019. THI was calculated from daily maximal temperatures and daily minimum humidity. Three sets of analysis were performed. In the first and second analysis, two mixed effect repeatability models were fitted with THI treated as a categorical variable grouped into 5 classes (THI1= [61 - 66], THI2= [67 - 71], THI3= [72 - 78], THI4=[79 - 81], THI5=[82 - 86]), to obtain least squares estimates of THI effect on milk production, and as a continuous variable within THI classes to identify THI thresholds at which milk yield started to decline. In the third analyses, one quadratic polynomial regression (POL) and three regression spline functions namely piecewise linear spline function (PLF), natural splines function (NSF) and cubic splines function (CSF) were fitted to determine the average effect of THI on milk yield in the population and describe the pattern of response of milk yield to increasing head load. The results show that heat stress reduced milk yield by 4.16% to 14.42% across THI groups, with daily milk yield being the highest in THI1 (7.40±0.39 litres) and the lowest in THI4 (6.33±0.32). Regression coefficients within groups showed significant daily milk yield decrease in THI2 (-0.09) and THI3 (-0.06), but not for other THI classes, indicating that cows experienced heat stress between THI values of 67 and 76 and milk loss plateaued afterwards, suggesting that the animals acclimatized to heat stress conditions beyond THI value of 76. At the population level, THI and its squared term were significantly negatively and positively (-0.61, 0.004) associated with milk production, indicating a non-linear relationship between milk yield and THI. The CSF model showed better goodness of fit and predictive ability than other models for predicting future population response of milk yield to heat stress in small holder dairy farms in Tanzania. Herd management strategies and animal husbandry measures are needed in small holder dairy farms in Tanzania to minimize the impact of heat stress on milk yield and income of the farmers.

Genomic Selection and Use of Molecular Tools in Breeding Programs for Indigenous and Crossbred Cattle in Developing Countries: Current Status and Future Prospects.

Genomic selection (GS) has resulted in rapid rates of genetic gains especially in dairy cattle in developed countries resulting in a higher proportion of genomically proven young bulls being used in breeding. This success has been undergirded by well-established conventional genetic evaluation systems. Here, the status of GS in terms of the structure of the reference and validation populations, response variables, genomic prediction models, validation methods, and imputation efficiency in breeding programs of developing countries, where smallholder systems predominate and the basic components for conventional breeding are mostly lacking is examined. Also, the application of genomic tools and identification of genome-wide signatures of selection is reviewed. The studies on genomic prediction in developing countries are mostly in dairy and beef cattle usually with small reference populations (500-3,000 animals) and are mostly cows. The input variables tended to be pre-corrected phenotypic records and the small reference populations has made implementation of various Bayesian methods feasible in addition to GBLUP. Multi-trait single-step has been used to incorporate genomic information from foreign bulls, thus GS in developing countries would benefit from collaborations with developed countries, as many dairy sires used are from developed countries where they may have been genotyped and phenotyped. Cross validation approaches have been implemented in most studies resulting in accuracies of 0.20-0.60. Genotyping animals with a mixture of HD and LD chips, followed by imputation to the HD have been implemented with imputation accuracies of 0.74-0.99 reported. This increases the prospects of reducing genotyping costs and hence the cost-effectiveness of GS. Next-generation sequencing and associated technologies have allowed the determination of breed composition, parent verification, genome diversity, and genome-wide selection sweeps. This information can be incorporated into breeding programs aiming to utilize GS. Cost-effective GS in beef cattle in developing countries may involve usage of reproductive technologies (AI and in-vitro fertilization) to efficiently propagate superior genetics from the genomics pipeline. For dairy cattle, sexed semen of genomically proven young bulls could substantially improve profitability thus increase prospects of small holder farmers buying-in into genomic breeding programs.

Use of body linear measurements to estimate liveweight of crossbred dairy cattle in smallholder farms in Kenya.

Body linear measurements, and specifically heart girth (HG), have been shown to be useful predictors of cattle liveweight. To test the accuracy of body linear measurements for predicting liveweight, crossbred dairy cattle of different genotypes were measured and weighed. A total of 352 mature cows and 100 heifers were weighed using an electronic weighing scale and measurements of HG, body length, height at withers were taken using an ordinary measuring tape and body condition scored (BCS) using a five-point scale. The animals were grouped according to genotype and age. Genotype classification was undertaken from farmer recall and by visual appraisal as 40-60, 61-80 or 81-100 % exotic (non-indigenous). Age classification was simply as mature cows or heifers. Liveweight of the animals ranged from 102 to 433 kg. Liveweight was strongly correlated with HG (r = 0.84) and body condition scores (r = 0.70) and moderately correlated with body length (r = 0.64) and height at withers (0.61). Regressing LW on HG measurements gave statistically significant (P < 0.01) equations with R(2) ranging from of 0.53 to 0.78 and residual standard deviation ranging from 18.11 to 40.50 kg. The overall model developed (adjusted R(2) = 0.71) had a prediction error of 26 kg (or 11 % of the mean) and predicted LW of over 95 % of crossbred dairy cattle in the range of 100-450 kg, regardless of age and breed group. Including BCS in the model slightly improved the model fit but not the prediction error. It was concluded that the model can be useful in making general management decisions in smallholder farms.

Identification of smallholder farmers and pastoralists' preferences for sheep breeding traits: choice model approach.

Identification of breeding objective traits pertinent to specific production environments with the involvement of target beneficiaries is crucial to the success of a breed improvement program. A choice experiment was conducted in four locations representing different production systems and agro-ecologies that are habitat to four indigenous sheep breeds (Afar, Bonga, Horro and Menz) of Ethiopia with the objective of identifying farmers'/pastoralists' preferences for sheep breeding traits. Following a synthesis of secondary information and diagnostic surveys, two communities per location consisting of 60 households each having at least four breeding ewes were identified. Producers' priority attributes used in the choice sets were identified through in-depth production system studies conducted from December 2007 to March 2008. On the basis of prior information, four to seven attributes were used to design choice sets with different profiles in order to capture results that mimic real life of the different communities. The attributes and levels chosen for the sheep profile were as follows: body size (large/small), coat color (brown/white/black), tail type (good/bad) for both rams and ewes; horn (polled/horned) and libido (active/poor) for rams; and lambing interval (three lambings in 2 years/two lambings in 2 years time), mothering ability (good mother/bad mother), twinning rate (twin bearer/single bearer) and milk yield (two cups per milking/one cup per milking) for ewes. A fractional factorial design was implemented to construct the alternatives included in the choice sets. The design resulted in a randomized selection of 48 sheep profiles (24 sets) for both sexes, which were grouped into four blocks with six choice sets each. An individual respondent was presented with one of the four blocks to make his/her choices. Results indicate that producers' trait preferences were heterogeneous except for body size in rams and mothering ability in ewes where nearly homogeneous preferences were investigated. In the pastoral production system, attention was given to coat color of both breeding rams and ewes, favoring brown and white colors over black. Ram libido influenced producers' decisions in Bonga, Horro and Menz areas. The influence of milk yield and twinning on respondents' decision making was high in Afar and Horro, respectively. Breeders in all areas attempt to combine production and reproduction traits as well as they can in order to maximize benefits from their sheep. The elicited measurable objective traits were used to design alternative community-based sheep breeding plans for the four indigenous sheep breeds in their production environments that have been implemented since.

Analysis of early survival of Holstein-Friesian heifers of diverse sire origins on commercial dairy farms in Kenya.

The use of imported semen within the Holstein-Friesian cattle population in Kenya has contributed to increased milk production per cow, however, information on how this has impacted on functional traits, particularly early life survival and reproductive performance is scarce. This study evaluated age at first calving (AFC), survival to age at first calving, and survival to four years of age using survival analyses techniques, in Holstein-Friesian cattle on four dairy farms in Kenya. The heritability estimate obtained for AFC was 0.15 +/- 0.06 for an average AFC of 1058 days. Animals sired by New Zealand and Australian born bulls had the earliest average AFC (907 days). On average, 25% of all the heifers born were culled prior to attaining a first calving, while 34% were culled prior to four years of age. Though the highest proportion of losses was due to unspecified reasons, the relative risk of being culled was highest when an animal had a specific disease, and the first 60 days of life were the most critical for survival. Daughters of sires from South-Africa and Israel tended to have better survival rates than those sired by bulls originating from other regions. Unfavourable selection towards animals sired by Kenyan born bulls was evident. The economic implications of the high rate of early mortality need to be evaluated in order to assist livestock producers make informed decisions on choice of sires for breeding.

The small east african shorthorn zebu cows in Kenya. I: linear body measurements.

Assessment of genetic diversity is a prerequisite for the management and conservation of animal genetic resources. Appropriate design of breeding programmes is therefore impossible for breeds that have not been adequately characterized either phenotypically and/or genetically. Phenotypic characteristics are important in breed identification and classification in ways that farming communities can relate with. This study phenotypically characterized two breeds of zebu cattle in Kenya. A total of 12 measurements (face length, ear length, horn length, heart girth, height at withers, chest depth, body length, height at rump, pelvis width, corpus length, pin bone width and tail length) were collected on 373 Maasai and 277 Kamba zebu kept by traditional farmers in south-east Kenya. The data were classified on the basis of breed group, age group, sex and coat colour pattern. Breed group, age group and sex significantly influenced all measurements. Coat colour pattern significantly influenced only height at withers, corpus length, ear length and tail length. Except for horn and ear length, all the other measurements were significantly higher for the Maasai zebu. Additionally, the Maasai zebu was taller than it was long. The opposite was true for the Kamba zebu. The Maasai and Kamba zebus can be classified as medium-sized breeds; however, great variations exist in their body sizes within and between the breeds.

Comparison of production systems and selection criteria of Ankole cattle by breeders in Burundi, Rwanda, Tanzania and Uganda.

A survey in Burundi, Rwanda, Tanzania and Uganda was conducted in order to determine the different production systems under which Ankole cattle are currently kept. Additionally, selection criteria of livestock keepers were documented. In Burundi, Rwanda and parts of Uganda, livestock keepers are sedentary and herds are small, whereas in the other areas Ankole cattle are kept in large herds, some of them still under a (semi-)nomadic system. Milk is the main product in all areas, and is partly for home consumption and partly for sale. Although the production systems vary in many aspects, the selection criteria for cows are similar. Productive traits such as milk yield, fertility and body size were ranked highly. For bulls, the trait 'growth' was ranked highly in all study areas. Phenotypic features (coat colour, horn shape and size) and ancestral information are more important in bulls than in cows. The only adaptive trait mentioned by livestock keepers was disease resistance. In areas of land scarcity (Burundi, Rwanda, western Uganda), a clear trend from pure Ankole cattle towards cross-bred animals can be observed.

Characterization of indigenous fat-tailed and fat-rumped hair sheep in Kenya: diversity in blood proteins.

This paper reports the variation in five blood proteins from five populations of sheep found in Kenya. Blood samples were collected from a total of 309 adult sheep of both sexes in Kwale, Makueni and Kakamega districts for the fat-tailed sheep and in Isiolo district for the fat-rumped hair sheep. Fine-wooled Merino sheep were used in this study as the reference population. Transferrin, esterase-A and esterase-C were polymorphic in all the populations investigated, while albumin was monomorphic for the S allele in the fat-tailed sheep and haemoglobin was fixed for the B allele in the Kwale, Makueni and Isiolo populations. Phylogenies derived from the pairwise genetic distance estimates showed a clear separation between the indigenous sheep populations and the exotic Merino. However, the topology of the former showed rather poor consistency with their morphological classification based on the localization of their fat deposits, namely fat-tailed or fat-rumped hair sheep.