Search

International audience; The economic importance of genetically improving feed efficiency has been recognized in dairy production. Feed efficiency is a genetically complex trait that can be described as units of product output (e.g. milk yield) per unit of feed input. Previous work showed that feed efficiency is heritable in many species. This study aimed to investigate the heritability for Residual Energy Intake in two breeds of dairy goats (Alpine and Saanen) under commercial conditions. Heritabilities were moderate for both breeds (0.18±0.08 in Alpine and 0.20±0.07 in Saanen).

International audience; The resource allocation theory has long been of interest in livestock as a framework to explain the correlated responses to selection for high productivity. Increasing limitations on feed resources and genetic improvement of feed efficiency (FE) should further promote allocation constraints. Yet, predicting those consequences for breeding critically depends upon the possibility to get testable predictions from the current framework. Here we assessed the changes in energy allocation when selecting for FE, and the consequences of those changes on health and reproduction. For this, we used two approaches; a survey of selection experiments among various species and a case study in sheep focusing on a potential trade-off between FE and response to an infectious challenge (parasite infection). Overall, results provide weak evidence for the current allocation framework. We outline several directions to better appraise the consequences of breeding for FE in increasingly challenging environments.

Industrial livestock systems faces a major crisis of legitimacy and animal breeding contributes to this situation. Worldwide, breeding goals focus on economic performance and rarely consider the environmental and social impacts. Agroecology offers a conceptual framework to think alternative solutions to face this crisis and to guide the transition of livestock farming towards nature-based systems. Animal genetics, as an essential component of animal husbandry, should contribute to this evolution. Based on the agroecological principles applied to animal farming and the efficiency-substitution-redesign (ESR) framework, we showed that animal breeding mainly focused on the E-S levels of the ESR framework and few on the redesign of animal farming. We then present some perspectives for the redesign of breeding goals, the use of the G×E interaction information in the evaluation of animals, and the benefits of managing the available genetic diversity of livestock species to fit to the needs of nature-inclusive systems.

International audience; Livestock production systems have considerably evolved over the 20th century. Research in animal breeding and genetics and the implementation of genetic improvement programmes have played an important role in this evolution. Today, the dominant model, characterized by an intensive use of inputs, a very high degree of specialization of production systems and the search for ever lower production costs, is questioned. A now widely shared objective is to contribute to the emergence of sustainable, equitable, healthy and environmentally-friendly food systems. Agroecology is a means to achieve this goal and a guide to the necessary transition of livestock systems, to which animal genetics must contribute. Examples of past, current and potential contributions are presented and positioned according to five agroecological principles proposed as a guide to the evolution of livestock systems. Most of them, such as the selection of animals resistant to different infectious diseases or making a more efficient use of feed, correspond to low levels of agroecological transitions in that they do not question the foundations, components or general design of production systems. Further contributions aimed at a strong transition, based on an in-depth redesign of livestock systems, should be developed in the future.; Les filières et systèmes d’élevage ont considérablement évolué au cours du XXe siècle. La recherche en génétique animale et la mise en place des programmes d’amélioration génétique ont joué un rôle important dans cette évolution. Aujourd’hui, le modèle dominant, caractérisé par une utilisation intensive d’intrants, une très grande spécialisation des systèmes et la recherche de coûts de production toujours plus bas, est remis en cause. Un objectif désormais largement partagé est de contribuer à l’émergence de systèmes alimentaires durables, équitables, sains et respectueux de l’environnement. L’agroécologie est un moyen pour atteindre cet objectif et guider la nécessaire transition des systèmes d’élevage, à laquelle la génétique animale doit contribuer. Des exemples de contributions passées, actuelles et potentielles sont présentés et positionnés selon cinq principes d’agroécologie proposés pour guider l’évolution des systèmes d’élevage. La plupart, telles que la sélection d’animaux résistants à différentes maladies infectieuses ou valorisant de façon plus efficace leur alimentation, correspondent à des niveaux de transition agroécologique faible, dans la mesure où elles ne remettent pas en cause les fondements, les composantes ou la conception générale des systèmes. De nouvelles contributions, visant une transition forte, fondée sur une reconception en profondeur des systèmes d’élevage, sont à développer à l’avenir.

Additional file 9: Figure S5. Sex assignment for the 1159 goats sampled in VarGoats. Representation of the ratio of read number between the autosomes and the X chromosome and the read number on the Y chromosome for each animal to determine its sex. The horizontal bar corresponds to the threshold of 25 reads from the Y scaffold allowing to differentiate males from females.

The enhanced availability of sequence data in livestock provides an opportunity for more accurate predictions in routine genomic evaluations. Such evaluations would therefore no longer rely only on the linkage disequilibrium between a chip marker and the causal mutation. The objective of this study was to assess the usefulness of sequence data in Saanen goats (n = 33) to better capture a quantitative trait locus (QTL) on chromosome 19 (CHI19) and improve the accuracy of predictions for 3 milk production traits, 5 type traits, and somatic cell scores. All 1,207 50K genotypes were imputed to the sequence level. Four scenarios, each using a subset of CHI19 imputed variants, were then tested. Sequence-derived information included all CHI19 variants (529,576), all variants in the QTL region (22,269), 178 variants selected in the QTL region and added to an updated chip, or 178 randomly selected variants on CHI19. Two genomic evaluation models were applied: single-step genomic BLUP and weighted single-step genomic BLUP. All scenarios were compared with single-step genomic BLUP using 50K genotypes. Best overall results were obtained using single-step genomic BLUP on 50K genotypes completed with all variants in the QTL region of chromosome 19 (6.2% average increase in accuracy for 9 traits) with the highest accuracy gain for fat yield (17.9%), significant increases for milk (13.7%) and protein yields (12.5%), and type traits associated with CHI19. Despite its association with the QTL region of chromosome 19, the somatic cell score showed decreased accuracy in every alternative scenario. Using all CHI19 variants led to an overall decrease of 4.8% in prediction accuracy. The updated chip was efficient and improved genomic evaluations by 3.1 to 6.4% on average, depending on the scenario. Indeed, information from only a few carefully selected variants increased accuracies for traits of interest when used in a single-step genomic BLUP model. In conclusion, using QTL region variants imputed from sequence data in single-step genomic evaluations represents a promising perspective for such evaluations in dairy goats. Furthermore, using only a limited number of selected variants in QTL regions, as available on SNP chip updates, significantly increases the accuracy for QTL-associated traits without deteriorating the evaluation accuracy for other traits. The latter approach is interesting, as it avoids time-consuming imputation and data formatting processes and provides reliable genotypes.

International audience

Additional file 6: Figure S4. Tranche plot produced by VariantRecalibrator for SNPs. Description: Partition of the call sets into quality tranches. The tranches correspond to certain levels of sensitivity relative to the truth sets (the highest tranche corresponds to a high accuracy call set but with the lowest value of sensitivity).

Additional file 3: Figure S1. Distribution of the concordance rates. Concordance rates (CR) between sequence variants and 50 k genotypes for 457 individuals. The figure shows a clear disruption in the distribution of CR, thus we found it easy and relevant to discard the samples below a concordance rate of 70% between sequence and chip SNP data.

Additional file 5: Figure S3. Pairwise comparisons of the annotations chosen in VQSR for InDels. Description: Modeling report generated by GATK VariantRecalibrator for every pairwise combination of annotations used (QD, DP, FS, MQRankSum, ReadPosRankSum and SOR), with a 2D projection of the Gaussian mixture model.

International audience

International audience

International audience; The development of statistical methods aiming to improve the accuracy of genomic predictions is of utmost value for dairy goat breeding programs. In this context, the use of haplotypes, instead of individual SNP, could improve the accuracy of genomic predictions by better capturing the effect of causal variants, instead of relying solely on linkage disequilibrium with individual SNP. Haplotypes can be included in genomic evaluation models in various ways, such as fitting them as pseudo-SNP (i.e., haplotypes converted into biallelic SNP format). This can be easily incorporated in the software already available for single-step genomic predictions (ssGBLUP). Therefore, the aim of this study was to compare the predictive performances of ssGB-LUP and weighted ssGBLUP (WssGBLUP) based on individual SNP or on haplotypes fitted as pseudo-SNP. Performance was compared in terms of accuracy, bias, and weights for SNP versus pseudo-SNP. Genomic predictions were performed on 5 milk production traits, 5 udder type traits, and somatic cell score (SCS). The training population was formed by 307 Alpine and 247 Saanen progeny-tested bucks, genotyped using the Il-lumina Goat SNP50 BeadChip (Illumina, San Diego, CA). The validation population included 205 Alpine and 146 Saanen young bucks. The accuracy of genomic predictions was evaluated in the validation population as the Pearson correlation between genomic estimated breeding values (GEBV), predicted based on various methods, and daughter deviation (DD) based on the official genetic evaluation of January 2016. Haplotype-based models were shown to improve the performance of genomic predictions for some traits. Gains in accuracy of up to +19% (0.310 to 0.368 for fat yield) in Alpine and up to +3% (0.361 to 0.373 for udder shape) in Saanen were observed with ssGBLUP. The ssGB-LUP accuracies averaged across all traits and methods were equal to 0.467 (SNP) versus 0.471 (pseudo-SNP) in Alpine and 0.528 (SNP) versus 0.523 (pseudo-SNP) in Saanen. With WssGBLUP, gains in accuracy of up to 24% (0.298 to 0.370 for fat yield) in Alpine and 14% (0.431 to 0.490 for SCS) in Saanen were observed with WssGBLUP. Accuracies of WssGBLUP averaged across all traits and methods were equal to 0.455 (SNP and pseudo-SNP) in Alpine and 0.542 (SNP) versus 0.528 (pseudo-SNP) in Saanen. The average (±SD) slope of the regression of DD on GEBV for the validation animals, across all breeds, traits and scenarios, were equal to 0.82 ± 0.20 (SNP) and 0.83 ± 0.18 (pseudo-SNP) for ssGBLUP and 0.67 ± 0.16 (SNP) and 0.65 ± 0.16 (pseudo-SNP) for WssGBLUP, which suggest that haplotype-based models and ssGBLUP SNP were similarly biased. However, WssGBLUP was more biased than ssGBLUP, and its gains in accuracies were limited to milk production traits. Despite the fact that genomic predictions based on haplotypes require additional steps and time, the observed gains in GEBV predictive performance indicate that haplotype-based methods could be recommended for some traits.

International audience; The main French dairy goats breeds Alpine and Saanen are selected using a total merit that combines milk production traits (protein and fat yields and protein and fat contents), udder morphology and udder health. Genomic selection has been implemented, thanks to the following steps of research: i) the sequencing of goat genome in 2010, ii) the creation of the IGGC to promote and coordinate the development of a high-density SNP ship enabling the design of the Illumina 50K SNP ship in 2011, iii) the implementation of a reference male population. Specificities of French dairy goats - small reference population size and low linkage disequilibrium – required a suitable statistical model. INRA studies showed that with a Single Step GBLUP, expected gains in accuracy would be of 0.06 for milk yield and 0.14 for fat and protein contents. These results encouraged INRA to implement genomic evaluations, and first GEBVs were published in 2018.The breeding scheme has been entirely renewed by the breeding organization Capgenes, particularly for AI bucks. The male selection intensity increased and the generation interval was reduced, thanks to the early choice of bucks among a larger number males born from planned mating and genotyped at one month of age, and thanks to the use of young bucks for AI. ). Genetic evaluation programs have been adapted to produce early GEBVs as soon as genotyping are available. With these improvements, an increase of genetic progress is expected (20 to 30%) in the coming years, resulting in a rise on the net margin of the breeders. Moreover, new traits can be introduced in breeding goals while maintaining the same efficiency on historical traits. GEBVs for female AI fertility and for semen production are expected for 2020 (for internal use in a first time), and studies are ongoing on longevity, milk persistency and feed efficiency

Additional file 4:Affected sub-functions of lipid metabolism during different responses to mastitis infection. Five most significant sub-functions of lipid metabolism that are altered during (I) overall, (II) early stage, (III) late stage, and (IV) cattle-specific responses. The results were obtained by IPA using the lists of significantly affected genes for each specific response. The sub-functions of the lipid metabolism are listed from the lowest to the highest p-value, and are reported with the involved genes. (DOC 64 KB)

Additional file 7:Affected molecular and cellular functions of the most dissimilar genes between E. coli and S. aureus . Five most significant molecular and cellular functions identified with IPA using the 34 most dissimilar genes between E. coli and S. aureus infections in cattle in vivo (experiment 1A, 1B, and 1C), as found with the PAMR software (Table 3). The identified molecular and cellular functions are listed from the lowest to the highest p-value, and are reported with the involved genes. (DOC 40 KB)

Additional file 5:Supplemental Figure S1 - Relationship between XBP1 and additional affected genes during the early stage response to mastitis. Gene network showing the connections, as identified with the IPA option "building pathways", between the gene XBP1 and other affected genes during (II) early stage response to mastitis infection. A. XBP1 is related and linked to several other affected genes. B. XBP1 is directly linked to the genes COPZ1, DDOST, KDELR2, KDELR3, RPN1, SEC23B, SEC24D, SEC61A1, and SRPR, as well as to genes of the proteasome and the MHC Class II complex. Supplemental Figure S2 - Relationship between SREBF1 and additional affected genes during the late stage response to mastitis. Gene network showing the connections, as identified with the IPA option "building pathways", between affected genes involved in lipid metabolism during (III) late stage response to mastitis infection. The gene SREBF1 seems to play an important role and is directly linked to other affected genes (violet colour), i.e. TRAF3IP3, CD36, SCD, SOD1, IDH1, THRB, RETN, PMVK, DBI, UCP2, HBS1, SC4MOL, and CYP27A1. Supplemental Figure S3 - Venn diagram showing the number of common and experiment-specific affected genes between (IV) cattle-specific response and the individual experiments 1A time point {3} and 2 time point {9}. Venn diagram illustrating the number of significantly affected genes in common (25) and distinct for the (IV) cattle-specific response (red: 421 genes), experiment 1A time point {3} (green: 745 genes), and experiment 2 time point {9} (blue: 55 genes). The lists of corresponding genes can be found in [Additional file 6]. The list of experiments and time points can be found in Table 1 and the list of meta-analysis combinations in Table 2. (DOC 1 MB)

Additional file of Strengthening insights into host responses to mastitis infection in ruminants by combining heterogeneous microarray data sources

International audience

International audience; Genetic selection focused purely on production traits has proven very successful in improving the productive performance of livestock. However, heightened environmental and infectious disease challenges have raised the need to also improve the resilience of animals to such external stressors, as well as their efficiency in utilising available resources. A better understanding of the relationship between efficiency and production and health traits is needed to properly account for it in breeding programmes and to produce animals that can maintain high production performance in a range of environmental conditions with minimal environmental footprint. The aim of this study was to perform a meta-analysis of genetic parameters for production, efficiency and health traits in sheep and goats. The dataset comprised 963 estimates of heritability and 572 genetic correlations collated from 162 published studies. A threelevel meta-analysis model was fitted. Pooled heritability estimates for milk production traits ranged between 0.27 ± 0.03 and 0.48 ± 0.13 in dairy goats and between 0.21 ± 0.06 and 0.33 ± 0.07 in dairy sheep. In meat sheep, the heritability of efficiency traits ranged from 0.09 ± 0.02 (prolificacy) up to 0.32 ± 0.14 (residual feed intake). For health traits, pooled heritability was 0.07 ± 0.01 (faecal egg count) and 0.21 ± 0.01 (somatic cell score) in dairy goats and 0.14 ± 0.04 (faecal egg count) and 0.13 ± 0.02 (somatic cell score) in dairy sheep. In meat sheep, the heritability of disease resistance and survival traits ranged between 0.07 ± 0.02 (mastitis) and 0.50 ± 0.10 (breech strike). Pooled estimates of genetic correlations between resilience and efficiency traits in dairy goats were not significantly different from zero with the exception of somatic cell score and fat content (À0.19 ± 0.01). In dairy sheep, only the unfavourable genetic correlation between somatic cell score and protein content (0.12 ± 0.03) was statistically significant. In meat sheep only, the correlations between growth and faecal egg count (À0.28 ± 0.11) as well as between growth and dagginess (À0.33 ± 0.13) were statistically significant and favourable. Results of this metaanalysis provide evidence of genetic antagonism between production and health in dairy sheep and goats. This was not observed in meat sheep where most of the pooled estimates had high standard errors and were non-significant. Based on the obtained results, it seems feasible to simultaneously improve efficiency and health in addition to production by including the different types of traits in the breeding goal. However, a better understanding of potential trade-offs between these traits would be beneficial. Particularly, more studies focused on reproduction and resilience traits linked to the animal's multitrait response to challenges are required.

International audience; Goats were domesticated 10,500 years ago with the aim of supplying milk, meat and fibres. Since then, breeds have specialized and adapted to their local environment developing specific genetic profiles. The VarGoats project is an international resequencing program, which aims at covering at best the genetic diversity of the Capra species. To date the sequence data 829 Capra hircus of various breeds and geographical origins. Variant calling led to the identification of a total of 110,193,942 variants on the 29 autosomal chromosomes. Quality checks were applied to sequence variants using various indicators: quality, depth, minor allele frequency and position. For French Alpine and Saanen breeds the concordance with 50k genotypes (CaprineSNP50 BeadChip) was checked. Mean concordance rate was 98.22% and ranged from 94.00 to 99.96% among individuals. Imputation was tested on the 23,338,436 filtered variants using FImpute software. Pedigree was provided and imputation was performed in a within-breed leaveone-out scenario. Imputation quality was checked for 4 individuals on all chromosomes and for every sequenced animal on chromosome 29. Mean concordance rate for chromosome 29 ranged from 73.49 to 85.71% and from 72.72 to 79.27% in Alpine and Saanen breeds respectively. On all chromosomes, average correlation between true and imputed sequence were 0.77 and 0.76 in Alpine and Saanen respectively. Imputation was then applied to a population of 1,129 French Alpine and Saanen males with at least 10 daughters phenotypes. Association analysis was performed on production and functional traits using polygenic models with the GCTA software. Using genotype data imputed to sequence level allowed to improve the significance of QTL previously identified with 50k SNP-chip only. In Saanen breed, 1.6 Mb region on chromosome 19 was identified as significantly linked to 4 conformation traits, 3 semen production traits, somatic cell count, milk yield and protein and fat contents. These researches provided insights on how to implement a quality check, imputation that will ensure the quality of subsequent analyses, and on new SNPs to be added to the 50k SNP-chip.

International audience

International audience; Goats were domesticated 10,500 years ago with the aim of supplying milk, meat and fibers. Since then, breeds have specialized and adapted to their local environment developing specific genetic profiles. The VarGoats project is an international resequencing program which aims at covering at best the genetic diversity of the Capra species. To date the sequence data available include 16 wild types and 578 Capra hircus of 65 different breeds. Variant calling led to the identification of a total of 105,772,894 variants on the 29 autosomal chromosomes. For French Alpine and Saanen individuals the concordance with 50k genotypes was checked. Mean concordance rate was 97.74% and ranged from 90.16% to 99.96%. Quality checks were applied to sequence variants using various indicators: QUAL, GQ, DP, MAF and position. Imputation was tested on the 5,926,620 filtered variants using FImpute software. Pedigree was provided and different reference populations were compared in a leave-one-out scenario. Mean concordance rate ranged from 90.6 to 92.5% and from 89.7 to 91.7% in Alpine and Saanen breeds respectively. Correlation between true and imputed sequence ranged from 0.78 to 0.83 and from 0.76 to 0.81 in Alpine and Saanen respectively. Imputation will be applied to a population of 3,618 French Alpine and Saanen goats genotyped with the CaprineSNP50 BeadChip. The next step will be to compare association studies using 50K data or SNPs imputed from the sequence data. This research provides insights on how to implement a solid quality check, and imputation that will ensure the quality of subsequent analyses.

International audience; Milk somatic cell count (SCC) is commonly higher in goats than in cattle and sheep. Furthermore, the ability of milk SCC to predict mastitis is considered lower in goats than in cattle and sheep, and the relevance of somatic cell score (SCS)-based selection in this species has been questioned. To address this issue, we created 2 divergent lines of Alpine goats using artificially inseminated bucks with extreme estimated breeding values for SCS. A total of 287 goats, 158 in high- and 129 in low-SCS lines, were scrutinized for mastitis infections. We subjected 2,688 milk samples to conventional bacteriological analyses on agarose and bacterial counts were estimated for positive samples. The SCS, milk yield, fat content, and protein content were recorded every 3 wk. Clinical mastitis was systematically noted. A subset of 40 goats (20 from each line) was subsequently challenged with Haemonchus contortus and monitored for anemia (blood packed cell volume) and fecal egg counts to see if SCS-based selection had an indirect effect on resistance to gastrointestinal nematodes. Milk production traits, including milk quantity, fat content, and protein content, were similar in both goat lines. In contrast, the raw milk SCC almost doubled between the lines, with 1,542,000 versus 855,000 cells/mL in the high- and low-SCS lines, respectively. The difference in breeding value for SCS between lines was 1.65 genetic standard deviation equivalents. The Staphylococcus spp. most frequently isolated from milk were S. xylosus, S. caprae, S. epidermidis, and S. aureus. The frequency of positive bacteriology samples was significantly higher in the high-SCS line (49%) than in the low-SCS line (33%). The highest odds ratio was 3.49 (95% confidence interval: 11.95-6.25) for S. aureus. The distribution of bacterial species in positive samples between lines was comparable. The average quantity of bacteria in positive samples was also significantly higher in high-SCS goats (69 ± 80 growing colonies) than in low-SCS goats (38 ± 62 growing colonies). Clinical cases were rare and equally distributed between high- (n = 4; 2.5%) and low-SCS (n = 3; 2.3%) lines. Furthermore, the larger the amounts of bacteria in milk the higher the SCS level. Conversely, goats with repeatedly culture-negative udders exhibited the lowest SCC levels, with an average of below 300,000 cells/mL. We therefore confirmed that SCS is a relevant predictor of intramammary infection and hygienic quality of milk in goats and can be used for prophylactic purposes. After challenge with H. contortus, goats were anemic with high fecal egg counts but we found no difference between the genetic lines. This result provides initial evidence that resistance to mastitis or to gastrointestinal nematodes infections is under independent genetic regulation. Altogether, this monitoring of the goat lines indicated that SCS-based selection helps to improve udder health by decreasing milk cell counts and reducing the incidence of infection and related bacterial shedding in milk. Selection for low SCC should not affect a goat's ability to cope with gastrointestinal nematodes.

Figure S1 Visualization of linkage disequilibrium (r2 × 100) between the 40 markers closest to the SOCS2 point mutation (rs868996547, in the middle). Figure S2 Visualization of linkage disequilibrium measured as squared correlation coefficient (r2) according to distance between markers on the 50 K ovine SNP chip. Figure S3 Components estimations according to the different models. One-trait methods correspond to eqs. (1) and (2) and two-traits methods to eqs. (3) and (4). Figure S4 Manhattan plots of estimated SNP effects using the best WssGBLUP approach for each phenotype (second iteration). On the left are presented analysis without the SOCS2 genotype among the markers and on the right, with the SOCS2 genotype (green point). Figure S5 Manhattan plots of estimated variance explained by 20 adjacent SNPs using the best WssGBLUP approach for each phenotype (second iteration). The horizontal red line represents the threshold of 1% adopted in this study. On the left are presented the analyses without the SOCS2 genotype among the markers and on the right, with the SOCS2 genotype. (DOCX 2190 kb)

National audience; Chez les petits ruminants comme chez les bovins, l’insémination animale (IA) est un outil puissant pour créer le progrès génétique au niveau d’un noyau de sélection et le diffuser dans les élevages. Le choix des futurs mâles producteurs de semence pour l’IA est donc stratégique. Or, le pourcentage de mâles éliminés avant même leur diffusion, pour des raisons de manque de précocité, de mauvaise production de semence, de défaut de morphologie fonctionnelle ou de standard, est important (autour de 50% en ovins et en caprins), ce qui a un impact direct sur l’efficacité technico-économique des schémas. Le projet CAS DAR Maxi’mâle a pour objectif d’optimiser la gestion des mâles reproducteurs afin d’augmenter l’efficacité des schémas de sélection. En particulier, il vise à anticiper les réformes subies, par l’identification précoce des mâles qui seront capables de donner de la semence de qualité, à faciliter l’entraînement et la collecte, et à faire progresser le niveau global de la population sur des caractères liés à la fonction sexuelle. Ce projet associe tous les centres de production de semence des petits ruminants français : 9 centres ovins et le centre caprin.

AdaptMap Consortium.

International audience

National audience

Economic sustainability of dairy ruminant farms mainly relies on production, reproduction, and health of livestock. In this study, we investigated the genomic regions controlling seventeen production traits (eleven type traits, milk yield, four milk composition traits and lifespan of livestock trait), five reproduction traits (semen production and quality), and one trait related to health (Somatic Cell Score, as a proxy for mastitis resistance). Daughter yield deviation values were computed for 597 (672) Alpine and 460 (519) Saanen AI rams from Capgènes testing station for production traits, mastitis (and reproduction). For lifespan, only the oldest rams had yield deviations to allow large progeny and good accuracy, i.e. 341 Alpine and 298 Saanen. All AI rams were genotyped with the Illumina GoatSNP50 Bead Chip. We conducted genome wide association studies using polygenic mixed models and genomic relatedness matrix with the GEMMA software. The two breeds were analysed separately. Results showed a highly significant QTL on CHI 19 (22.8-28.9 Mb) in the Saanen breed for 12 of 23 studied traits. The most significant traits were udder floor position (P=7.31e-24), volume of semen production (P=8.43e-21) and protein yield (P=6.64e-17). We did not find this QTL in the Alpine breed suggesting a breed-specific control. Analyses of sequencing data of 11 Alpine and 9 Saanen rams highlighted a candidate causal mutation in the Sex Hormone Binding Globulin (SHBG) gene. We will determine whether these pleiotropic relationships are further supported by one single mutation or resolved into closely linked loci. Altogether, these results identify a major pleiotropic QTL in the Saanen breed. They will be helpful for goat selection in the future. They indicate that a within-breed genomic selection should be favoured for those traits and that including QTL information might prove useful. The study further pave the way for the addition of reproduction traits in the French goat breeding programs.