Colostrum yield (CY, L) and concentration of immunoglobulin G (IgG, g/L) are important phenotypes to monitor in dairy farms because of their association with the risk of failure of the passive transfer of immunity in the newborn calf. This can occur when the CY of the parturient cow is insufficient or when the IgG concentration is low. Given that both of these traits are heritable, the present study aimed to investigate their genetic determinism by identifying significant genomic regions. A genome-wide association study coupled with an exploratory functional enrichment analysis was carried out to provide preliminary biological context of the detected signals for the ‘colostrability’ defined as the cow’s ability to secrete enough volume (≥ 4 L) of good-quality colostrum (≥ 50 g IgG/L) at calving. Data comprised 960 genotyped Italian Holstein cows with CY recorded within 6 h of calving, together with colostral IgG and total immunoglobulin concentrations. The significant SNPs associated with CY were scattered across BTA3, 5, 10, 11, 21, and 22, with 23 genes on BTA22 either harbouring or flanking significant signals. Apart from some genes already known, part of the significant regions have unclear function. Signals were detected on BTA1, 6, 7, 11, 19, 21, and 25 for IgG concentration, and on BTA6, 7, 11, 21, 23, and 25 for total immunoglobulin concentration. The functional enrichment analysis provided preliminary support for possible involvement of secretory, immune-signalling, and epithelial receptor-related processes. This study confirms the polygenic nature of cows’ ‘colostrability’ being regulated by different genomic regions distributed across the genome. However, exploration of the genomic determinism of CY and immunoglobulin concentration requires larger, independent, and harmonized data, ideally standardized and highly comparable. These findings, although relevant for improving calf health, represent only part of a more complex picture when the goal is selective breeding toward calf health. In addition to dam-related data, including colostrum traits, future studies should integrate calf-related phenotypes associated with failure of passive transfer of immunity, such as intestinal IgG absorption capacity, gut permeability, early-life survival, and health outcomes.
Body Size traits, particularly body weight (BW) and body mass index (BMI) at slaughter age, determine the meat yield and productivity of pigs. These phenotypes are shaped by numerous small-effect polygenes and regulated mostly by non-coding variants. Although genome-wide association studies (GWAS) have identified several loci and candidate functional variants, the regulatory mechanisms and causative genes of most traits remain uncharacterized, which limits the effectiveness of genomic prediction (GP). The purpose of this study was to bridge the gap between association studies and GP by integrating regulatory genomics into the GP framework to enhance prediction accuracy for body size traits. Using imputation-based GWAS in 1226 Shanxia Long Black pigs, multiple genome-wide significant loci were identified to be associated with BW and BMI. Linkage disequilibrium (LD) analysis, SuSiE fine-mapping, and regulatory modeling with Basenji deep-learning predictions refined these associations to 10 quantitative trait loci (QTLs) with 45 high-confidence candidate functional variants. Integration of chromatin-state annotations and high-throughput chromosome conformation capture (Hi-C) data revealed receptor tissue regulatory architectures; BW-associated variants on Sus scrofa chromosome 2 (SSC2) were enriched for brain regulatory regions, whereas BMI-associated loci showed enhancer activity across adipose, brain, and liver tissues. Multi-omics analyses converged on ZER1, KLHL29, and HAO1 as high-confidence candidate genes, while OR2T27 was a putative candidate on SSC2. In Basenji prediction, several specific candidate variants were identified as a liver enhancer. Incorporating these top-prioritized functional variants into genomic prediction models, GP yielded up to 21
Avian Influenza virus (AIV) and Newcastle disease virus (NDV) are highly contagious immunosuppressive pathogens that cause high mortality rates and economic losses in the global poultry industry. Although vaccination to enhance the host antibody response remains the main control strategy, the genetic basis of this variation is poorly understood. In this study, we used a genome-wide association study (GWAS) to investigate the genetic architecture of the longitudinal antibody responses to AIV (subtypes H5, H7, and H9) and NDV in chickens. To identify the genomic regions and candidate genes associated with the longitudinal antibody response to AIV and NDV after immunization, we conducted a GWAS using 1,359,927 SNP markers in an advanced intercross line (AIL) of the F19 resource population. We found moderate heritability for both the traits. Based on these associations, we identified the genomic regions and candidate genes for both traits. Association analysis revealed two significant SNPs located on chromosome 3, rs316474025 near NFKBIE and SLC35B2 [rs316474025], and rs312870707 near ATG5 gene, that are associated with viral antibody titers against AIV subtypes H5 and H7, respectively. Additionally, one variant on chromosome 27, rs318005864 near DDK12 and RPL19 genes, was strongly linked to the antibody response associated with NDV at mid-phase (40 days post-immunization). Furthermore, another variant, rs737164663, near FRMD5 and GLCE genes, was strongly linked to the antibody titer associated with both AIV subtype H9 and NDV in the late phase (60 days post-immunization). Genes located in these regions may be responsible for the immune response in chickens. Before implementation in breeding programs, the efficacy and long-term persistence of selecting these antibody response traits should be validated in larger independent cohorts. We confirmed the presence of genetic variability and identified SNPs significantly associated with antibody titer traits in F19 chickens. These findings highlight genomic regions contributing to variation in antibody responses and provide valuable information for improving antibody-related traits through selective breeding programs.
While epigenetic variations can contribute to shaping phenotypic diversity, it can be challenging to isolate and quantify the portion of trait variability under non-genetic influence. In this study we compared the phenotypic responses for different traits of two epilines of Japanese quails (Coturnix japonica) across three generations, using a large sample size. These epilines were built in parallel following (epi +), or not (epi -), an initial genistein ingestion in the ancestors' diet and were maintained to harbour a similar genetic structure. Linear models were fitted to extract the fraction of variance allocated to multiple factors such as family, sex and epiline. The latter was found to be significantly associated with body weight. The contribution of the epiline to phenotypic variability progressively increased from the first generation (G0) to the last (G2), leading—for example in body weight at slaughter—to an average difference for adult males and females in G2 of 9 g and 14 g respectively, between epi + offspring and controls (epi-). Although these findings suggest genetic drift, they could also reveal a possible transgenerational effect of the initial diet disruption. The analysis of other phenotypes displayed rare significant effects of the epiline. This innovative experimental design offered a unique opportunity to better understand the evolution of phenotypic variability and the parameters constituting it across three generations following or not an environmental change.
Labour dystocia (LAD) in sows is a common reproductive disorder that reduces piglet survival rates and increases the number of stillbirths, causing substantial economic losses to pig farms. However, the genetic basis of LAD remains elusive. Herein, we performed a single-breed genome-wide association study (GWAS) for LAD using imputed whole-genome sequence data from 3263 sows (487 Landrace and 2776 Yorkshire). In this study, analysis of the reproductive traits showed that the total number of piglets born and the number of piglets born alive were 1.43–2.85 and 1.60–2.94 higher, respectively, in sows with natural labour than in those with LAD. We identified 250 and 12 SNPs associated with LAD in primiparous and multiparous sows, respectively. Furthermore, in primiparous Yorkshire sows, two major QTL were fine-mapped to a 498.44 kb interval (23.18–23.68 Mb) on SSC2 and a 779.85 kb interval (43.34–44.12 Mb) on SSC9. Further analysis revealed that the two significant SNPs (2_23340612 and 9_44119733) within these QTL regions were located within regulatory regions across multiple pig breeds and tissues. The potentially regulatory SNP (2_23340612) is located within the binding sites of 27 transcription factors, that are primarily involved in oxytocin signalling, muscle contraction, and the development of female genitalia and embryo. The other potentially regulatory SNP (9_44119733) is located within the binding sites of eight transcription factors, among which three transcription factors (ZNF274, HMGA1, and CTCF) are predicted to interact with six candidate genes (APOA1, APOC3, APOA4, APOA5, BUD13, and ZPR1). Finally, eight promising candidate genes (APOA1, APOC3, APOA4, APOA5, BUD13, ZPR1, ACAN, and HAPLN3) were identified to be associated with LAD. Functional enrichment analysis indicated that the candidate genes were enriched in pathways related to lipid metabolism, extracellular matrix organization, cell adhesion and response to estrogen. The LAD is a prominent reproductive disorder trait that negatively impacts the reproductive efficiency of sows. This study identified potentially regulatory SNPs (2_23340612 and 9_44119733) and eight promising genes (APOA1, APOC3, APOA4, APOA5, BUD13, ZPR1, ACAN, HAPLN3) associated with LAD. To our knowledge, this is the first whole-genome-sequence-based GWAS with large-scale reproductive data to identify genetic markers and candidate genes of LAD of sows.
Dominance is a non-additive genetic effect involving allele interaction at the same locus. In animal breeding, it has often been ignored due to its complexity and the low accuracy of pedigree-based estimates. This study analysed dominance variance components for reproductive traits [age at first foaling (AFF), reproductive efficiency (RE), interval between first and second foaling (I12)] and morphological traits [scapulo-ischial length (SIL), dorsal sternal diameter (DSD), knee circumference (KC)] in Pura Raza Española (PRE) mares. The aim was to assess the magnitude of dominance effects across traits to determine whether alternative models are required for breeding value estimation or if dominance deviation predictions should be reconsidered in strategies such as mating design. Mares with full sisters and at least one foaling were identified and additional records of other females foaling in the same stud and year were also selected. The number of mares with reproductive and morphologic records was 3967 and the pedigree included 11,331 individuals with an average relatedness coefficient of 5
Indigenous sheep of Greece represent important genetic resources shaped by long-term adaptation to diverse environments. However, many populations face demographic decline, genetic erosion, and their genomic diversity remains inadequately characterized. This study presents a comprehensive genome-wide analysis of both recognized and previously uncharacterized indigenous Greek sheep. Newly genotyped data from 36 Greek breeds/populations, one Cypriot breed and one outgroup (Cypriot Mouflon) were combined with previously published genotypes from 85 international breeds and four additional outgroups, analyzing 127 ovine populations in total. To mitigate commercial BeadChip ascertainment bias when evaluating these uncharacterized populations, we utilized genome-wide SNP blocks to assess genetic diversity, reconstruct population structure, estimate effective population sizes, and place them within a broad comparative framework encompassing European, Southwest Asian, and North African breeds. Genome-wide analyses evaluating 46,733 SNPs and 4347 multi-allelic haplotype blocks revealed a distinct ascertainment bias affecting Western and Eastern breeds differently. Mitigating this bias via the block-based approach demonstrated that Greek sheep breeds/populations retain high genetic diversity, despite pronounced heterogeneity. Breeds such as Lesvos, Vlahiko, and Karagouniko exhibited high heterozygosity, low inbreeding, and relatively large effective population sizes, whereas Thraki, Agrinio, Katafygio, Serres, and Argos showed low diversity, elevated inbreeding, and small recent effective population sizes. Greek breeds/populations occupied an intermediate position between Western European and Southwest Asian groups, reflecting their geographic location and historical role in early dispersal routes. Island populations (Cretan breeds, Kasos and Karpathos) formed a cohesive genetic cluster shaped by long-term isolation, while semi-fat-tailed Greek breeds showed close affinities with Middle Eastern and North African populations. Introgression from East Friesian sheep strongly influenced the genomic profile of the Arta breed. This study provides a comprehensive genomic baseline for indigenous Greek and Cypriot sheep, highlighting their high genetic diversity and complex demographic histories. While several breeds represent valuable reservoirs of adaptive variation, others face immediate risk of genetic erosion. These findings provide essential guidance for prioritizing conservation actions and integrating genomic information into sustainable management and breeding strategies.
Resilience is an important breeding objective in livestock, reflecting the capacity of animals to maintain their performance or recover quickly from short-term environmental perturbations. The derivation of resilience indicators requires longitudinal measurements of key traits. In beef cattle, the limited availability of frequently recorded variables restricts the development of resilience-related indicators. This limitation is particularly relevant in (sub)tropical regions of the Southern Hemisphere, where animals are exposed to greater environmental variability and indicators of growth stability could therefore be especially valuable. Therefore, the main objective of this study was to derive resilience- and uniformity-related indicators from sparse body weight records in a Brazilian Angus–Brangus population, estimate their genetic parameters, and assess their associations with other economically relevant traits. Six indicators potentially related to growth resilience and uniformity were derived from observed and expected body weight trajectories, representing complementary biological dimensions: area under the curve, mean absolute deviation, logarithm of the mean of squares, logarithm of variance, relative maximum drawdown, and recovery slope index. A cubic quantile regression provided the best fit for reference growth trajectories compared with nonlinear models. All indicators were heritable, with direct and maternal heritability estimates ranging from 0.029 to 0.240 and from 0.024 to 0.071 for logarithm of variance and recovery slope index, respectively. Relative maximum drawdown (0.24) and recovery slope index (0.18) showed the highest heritabilities, while logarithm of the mean of squares (0.039) and logarithm of variance (0.029) had the lowest ones. The additive genetic coefficient of variation was highest for recovery slope index, indicating substantial genetic variability. Genetic and phenotypic correlations formed biologically consistent clusters, with area under the curve–mean absolute deviation and logarithm of the mean of squares–logarithm of variance closely related (> 0.90), and relative maximum drawdown–recovery slope index strongly associated (− 0.93) but distinct from the others. Partial genetic correlations, accounting for average growth potential, were modest but consistently favorable across 16 growth, carcass, adaptation, temperament, and reproduction traits. Mean absolute deviation showed favorable associations with key traits, including pre-weaning gain (− 0.39) and yearling conformation score (− 0.27), as well as with tick count (0.16). In contrast, the recovery slope index showed favorable correlations with pre- and post-weaning gain (0.39 and 0.32), backfat thickness (0.30), and age at first calving (− 0.15). Overall, the recovery slope index combined high heritability, favorable genetic relationships with key traits, and low redundancy with other indicators. Resilience- and uniformity-related indicators derived from sparse body-weight records of beef cattle are heritable with substantial additive genetic variance. Recovery slope index and mean absolute deviation are the most promising indicators for breeding applications. Recovery slope index captured variation in growth recovery not represented by other indicators and showed strong potential for improving growth recovery ability, while mean absolute deviation provided a broad and consistent profile of favorable associations with growth, carcass, adaptation, temperament, and reproduction traits. Overall, this study establishes a framework for quantifying resilience- and uniformity-related variation in growth trajectories of Angus–Brangus beef cattle, thereby contributing to productivity and sustainability under challenging environmental conditions.
Advancements in sequencing technologies have led to an unprecedented availability of whole-genome sequencing data in all life sciences, including livestock research. However, this raises concerns regarding the accuracy of the associated metadata, particularly information on an individual’s subspecies or breed. In this analysis, the 1000 Bull Genomes project was used as an example for a large-scale dataset with structured metadata. We applied a framework combining a query of the NCBI BioSamples database with principal component analysis, admixture analysis, and distance metrics based on the genomic information to assess metadata integrity. The main decrease in metadata quality results from missing breed assignments for 6
Phosphorus (P) and calcium (Ca) are essential minerals for laying hens. Phosphorus in plant feeds is mainly stored as phytate and needs to be released by the enzyme phytase. Due to the high requirement of Ca, laying hens exhibit limited endogenous phytate degradation and thus plant-P is available to a limited extent. Mineral P supplemented to laying hen feed reduces phytate degradation further and decreases myo-inositol release in the intestinal tract, which is known to have many functions in poultry metabolism. The focus of this study was the investigation of P and Ca metabolism in the peak period of egg production in commercial hybrid laying hens from Lohmann Selected Leghorn (LSL; n = 200) and Lohmann Brown (LB; n = 200) strains at the phenotypic and quantitative genetic level using data referring to blood plasma, ileal digesta, excreta, and eggs. Population genetic analyses revealed larger genetic diversity in LB than LSL and substantial differentiation between the strains. The majority of Ca and P metabolism traits differed significantly in trait mean or variance between the two strains. The LB strain showed more trait variation at the phenotypic and quantitative genetic levels. Moderate to high and significant heritabilities were estimated for myo-inositol in the plasma ( h^2 = 0.43 for LSL and h^2 = 0.36 for LB), ileum digesta ( h^2 = 0.60 for LB; not estimable for LSL) and egg ( h^2 = 0.69 for LSL and h^2 = 0.55 for LB), and for the Ca concentration in the plasma ( h^2 = 0.27 for LB; not estimable for LSL). Noticeably significant phenotypic correlations between the traits of P and Ca metabolism measured in excreta, plasma, and ileal digesta were present in both strains. The study provided a comprehensive insight into P and Ca metabolism under standardized conditions in the two commercial laying hen strains LSL and LB during egg laying. Differences between the strains were present at the phenotypic and quantitative genetic level. Thereby, the hens’ genetics appeared to be a relevant driver of P and Ca metabolism, with LB showing more variability. The study confirmed population genetic differences between the strains. Despite the detected strain differences, significant correlations among the traits of P and Ca metabolism indicate that the general relationships between traits are comparable in both strains.
The objective of this study was to test whether the negative effects of inbreeding on production traits varied according to the level of environmental load. Traits analyzed were milk, fat, and protein yields, and somatic cell score (SCS). Environmental conditions were described using temperature (TEMP), relative humidity (RH), and the temperature–humidity index (THI), each divided into five equally sized classes. For each trait, the environmental variable with the largest impact was used to evaluate inbreeding effects across its classes. Inbreeding was measured using pedigree (FPED), the diagonal of the genomic relationship matrix (FGRM), and runs of homozygosity (FROH). Genomic-based inbreeding measures resulted in larger estimated inbreeding effects, compared to pedigree measures, with FGRM and FROH showing similar results. RH most affected milk yield, with losses of 600 g/day in the highest RH class. Inbreeding led to losses of 100 g/day per 1
With the successful application of machine learning (ML) in many areas, its potential for genomic prediction (GP) in animal and plant breeding has attracted growing attention. However, despite numerous studies, the benefit of ML models over GBLUP remains controversial. Some studies reported higher accuracy of ML than linear models with small reference populations, but this benefit often disappeared in larger datasets. Additionally, the number of QTL also affects the prediction accuracy of ML models. The aim of this study was to compare the performance of three ML models, random forest (RF), support vector regression (SVR), and multilayer perceptron with residual networks (MLP-ResNet), with that of GBLUP. We simulated livestock populations with different data characteristics, including heritability, reference population size, and the number of QTL underlying the trait. Our results showed that the number of QTL strongly affected the prediction accuracy of RF and MLP-ResNet, but had little effect on GBLUP and SVR. The accuracy of RF dropped markedly with increasing QTL number, whereas the accuracy of MLP-ResNet decreased with increasing QTL number when reference populations exceeded 10,000. Both RF and MLP-ResNet achieved benefits over GBLUP only when the number of QTL was small (i.e. ≤ 100). RF outperformed GBLUP with small reference populations, whereas MLP-ResNet outperformed GBLUP with large populations. SVR showed marginally lower accuracy than GBLUP across scenarios and required more computation time. Among all models, GBLUP showed the lowest dispersion bias. RF and MLP-ResNet achieved higher accuracy than GBLUP only when the number of QTL was small, whereas in all other situations GBLUP outperformed the ML models. The benefit of ML models over GBLUP occurred mainly in scenarios where the infinitesimal model assumption of GBLUP did not hold. Our results suggest that when introducing a new GP model, particularly tree-based or neural network methods, it is essential to evaluate its performance on simulated datasets across different numbers of QTL and reference population sizes. These findings provide insights for applying ML in animal and plant breeding.
Val Rendena, an isolated Alpine valley in northern Italy, is home to an autochthonous, dual-purpose cattle breed with unique historical and morphological traits, which has been preserved by local breeders despite severe epidemics since the 1700s. While previous genome-wide studies identified signatures of selection in Rendena cattle, little is known about its evolutionary history. To address this issue, we analyzed complete mitogenomes from 137 Rendena individuals, selected to represent the majority of maternal lineages across the breed, as well as mitogenomes from 31 Alpine Grey individuals, purportedly closely related to Rendena cattle. We identified 86 distinct mitochondrial DNA (mtDNA) haplotypes in the Rendena breed, indicating a high haplotype diversity (Hd = 0.986). Phylogenetic analyses revealed that virtually all samples belong to the T macro-haplogroup (T3 = 91
Fertility is an important but often cryptic and intrinsic characteristic of domesticated animals. Predicting reproductive potential is of great importance for the industry but assessment through indirect proxies is laborious and often impractical. Among other biological factors, genetic effects are expected to play a crucial role in shaping male and female fertility. In cases where heritable components are strong, polygenic merit could be a valuable tool for decision-making in breeding schemes. Here we estimate sex-specific variance components affecting fertilization success by analyzing outcomes of over 3000 controlled mating events in an Arctic charr breeding nucleus from Iceland. Furthermore, a machine learning framework using relationships-to-founders vectors as input and a two-tower neural network architecture is proposed and tested for prediction of fertilization success. Both approaches seem to capture a meaningful biological signal and offer alternative tools for ranking, selecting or even allocating matings between breeding candidates.
Small effective population size and the disproportionately large use of few genetically superior bulls in artificial insemination lead to extensive runs of homozygosity and an increased risk of homozygosity for deleterious alleles in domestic cattle, which may cause inbreeding depression. The adverse effects of inbreeding on phenotypic performance are well established, but the genetic variants contributing to inbreeding depression remain largely unknown. This study aimed to analyse the impacts of inbreeding on stature (measured as height at the sacral bone) in a cohort of 15,306 Brown Swiss (BS) cows that have imputed genotypes at 20 million sequence variants and stature measurements as height at the sacral bone. The average genomic inbreeding coefficient of the 15,306 BS cows estimated from runs of homozygosity (ROH) was 0.369 (± 0.022). We found a loss in stature, with height at the sacral bone decreasing by 0.076 cm per 1
Feed efficiency is an economically important but costly trait to measure in pig breeding. Previous studies have shown that integrating metabolomic data, such as proton nuclear magnetic resonance (¹H NMR) - derived metabolomic profiles, into genomic prediction models can improve the accuracy of estimated breeding values (EBVs) - for example, using a univariate metabolomic-genomic best linear unbiased prediction (MGBLUP) model for malting quality traits in barley and for average daily gain (ADG) in pigs using NMR-based metabolomic features (MFs). In this study, we extend this approach to predict feed conversion ratio (FCR) in pigs. We tested two hypotheses: (1) incorporating NMR metabolomic data into a univariate MGBLUP model increases the accuracy of EBVs for FCR compared with a univariate genomic BLUP (GBLUP) model, and (2) a bivariate MGBLUP model that jointly analyses FCR and the correlated trait ADG further improves EBV accuracy compared with a univariate MGBLUP model. We tested these hypotheses using an offspring-validation design, allowing prediction of EBVs for animals lacking individual FCR records. The experimental population comprised 8,174 Duroc pigs (4,027 males from a test station and 4,147 females from breeding herds). To evaluate the accuracy of EBVs for FCR, males with FCR records were used as the training population, and females without FCR records served as the validation population. These validation females had offspring with recorded FCR, and EBV accuracy was assessed by correlating their EBVs with the corrected phenotypes of their offspring. Incorporating metabolomic data into the univariate MGBLUP model generated EBVs for FCR that were 2.3
Crossovers are crucial in meiosis and can break up the link between neighboring loci, creating novel allele combinations and genetic variation. However, their potential use in livestock breeding programs remains unexplored. Here, we studied the patterns of crossover count in two pig lines with different selection histories. We analyzed genomic data from over 47,000 individuals in a boar line and 78,000 individuals in a sow line, using over 20,000 autosomal single nucleotide polymorphisms (SNPs). We estimated and compared autosomal crossover counts (ACC), intrachromosomal genetic shuffling, and inter-crossover distances when two crossovers are transmitted on a chromosome. Finally, we compared the recombination landscapes of individuals transmitting a high versus a low number of crossovers. The mean ACC was 19.9 (SD: 4.7) for the boar line and 20.9 (SD: 5.0) for the sow line. On all chromosomes and in both lines, 0, 1 or 2 crossovers were transmitted in > 93
Germline de novo mutations (DNMs) are rare events in mammals, typically occurring only a few dozen times per generation. These mutations are not entirely random; several factors are known to influence their rate, including DNA methylation. In this study, we leveraged the unique population structure of cattle with a few ancestors having a large contribution to the current gene pool, along with comprehensive genomic resources, to investigate mutational processes. We applied two complementary approaches: (1) identifying DNMs accumulated over generations in family trio segments (from identical-by-descent segments between descendants and direct ascendants spanning several generations) in Holstein and Montbéliarde breeds, and (2) detecting rare bi-allelic substitution variants (Minor Allele Frequency < 0.001) from a large panel of sequenced Holstein animals. Overall, transitions were over-represented compared to transversions for both DNMs and rare substitutions. Considering the nucleotide context, a notable enrichment of C > T substitutions was observed within CpG sites (CpG > TpG). This enrichment was particularly pronounced in low CpG density regions and positively correlated with local methylation levels in both gametes and several somatic tissues. Additionally, several transposable elements exhibited higher mutation rates relative to the rest of the genome, particularly young SINE and LINE elements. Together, these results provide insights into the mutational landscape in cattle and reinforce observations made in other mammalian species.
Background:Sustainable breeding programs need to balance short-term genetic improvement with the conservation of genetic diversity. While genomic selection has considerably increased the genetic gain for many breeding programs, the consequences on diversity can be less desirable. This is particularly the case for rare alleles and de-novo mutations, as markers used in genomic selection are generally not strongly associated with rare alleles. Moreover, genomic selection allows for the selection of young individuals without records, thereby ignoring the effects of de-novo mutations. We aimed to evaluate various selection strategies in terms of long-term genetic gain and conservation of genetic variance, with a focus on the use and conservation of favorable rare alleles and de-novo mutations.Results:To study these selection strategies, we simulated populations of 1000 individuals subject to 50 generations of selection with a trait with only additive gene actions, a trait with additive and dominance gene actions, and a trait with additive, dominance and epistatic gene actions. For each trait, we evaluated five genomic selection strategies that balance between genetic improvement and diversity management, namely: truncation selection, which only focuses on short-term genetic gain; optimal contribution selection, which balances short-term genetic gain with a constraint on the relatedness of the selected individuals; two versions of allele-reweighted selection, which upscale the effect of rare alleles in the breeding values; and constrained allele loss selection, a novel strategy which balances short-term gain with a constraint on the reduction in frequency of rare alleles estimated to be favorable. Our results show that the allele-reweighted strategies provided an efficient trade-off between conserving genetic variance and achieving a higher genetic long-term gain, improving one or both metrics relative to truncation selection. Optimal contribution selection also improved the amount of genetic variance conserved and, for the trait with epistatic gene actions, also resulted in higher long-term genetic gain. On the other hand, the constrained allele loss did not show improvements over truncation selection.Conclusions:Introducing into our genomic selection strategies a consideration for diversity management or the conservation of rare alleles can help in improving the long-term sustainability of breeding programs that use genomic selection.
As a key index of reproductive performance in pigs, gestation length (GL) exerts a direct influence on litter size and the survival rate of piglets. Phenotypic variability in GL has been observed across Large White pig populations. To elucidate the genetic mechanisms underlying GL, this study analyzed 22,783 reproductive records from 9,057 pigs across five Large White populations. We employed a combination of diverse analytical strategies, encompassing genome-wide association studies (GWAS), GWAS meta-analysis, Bayesian fine-mapping (BFM), transcriptome-wide association studies (TWAS), phenome-wide association studies (PheWAS), and epigenetic profiling. GWAS and meta-analysis identified multiple novel GL-associated genomic regions on Sus scrofa chromosomes (SSC) 2, 7, 9, and 14. BFM refined the confidence intervals of these quantitative trait loci (QTLs), narrowing them to 10,460,892–10,737,692 bp and 31,384,086–31,384,333 bp on SSC7:, to 127,322,114–128,205,369 bp on SSC9:, and to 24,640,681–25,479,574 bp on SSC12:. TWAS revealed significant GL-related gene expression changes in relevant tissues, identifying GNAZ and SLC5A4—both located within the fine-mapped QTL on SSC14 (48.25–49.02 Mb)—as key regulators of GL. PheWAS further validated the association of GNAZ and SLC5A4 with GL, while epigenetic profiling showed active chromatin states and broad promoter activity at the GNAZ locus, supporting its regulatory potential. Through multi-omics integration, this study pinpoints GNAZ and SLC5A4 as core functional candidate genes regulating GL in Large White pigs. These results yield new perspectives on the genetic underpinnings governing porcine GL and supply valuable genetic materials for enhancing reproductive performance in swine breeding initiatives.
