Evaluation of low-density SNP panels and imputation for cost-effective genomic selection in four aquaculture species.

Genomic selection can accelerate genetic progress in aquaculture breeding programmes, particularly for traits measured on siblings of selection candidates. However, it is not widely implemented in most aquaculture species, and remains expensive due to high genotyping costs. Genotype imputation is a promising strategy that can reduce genotyping costs and facilitate the broader uptake of genomic selection in aquaculture breeding programmes. Genotype imputation can predict ungenotyped SNPs in populations genotyped at a low-density (LD), using a reference population genotyped at a high-density (HD). In this study, we used datasets of four aquaculture species (Atlantic salmon, turbot, common carp and Pacific oyster), phenotyped for different traits, to investigate the efficacy of genotype imputation for cost-effective genomic selection. The four datasets had been genotyped at HD, and eight LD panels (300-6,000 SNPs) were generated in silico. SNPs were selected to be: i) evenly distributed according to physical position ii) selected to minimise the linkage disequilibrium between adjacent SNPs or iii) randomly selected. Imputation was performed with three different software packages (AlphaImpute2, FImpute v.3 and findhap v.4). The results revealed that FImpute v.3 was faster and achieved higher imputation accuracies. Imputation accuracy increased with increasing panel density for both SNP selection methods, reaching correlations greater than 0.95 in the three fish species and 0.80 in Pacific oyster. In terms of genomic prediction accuracy, the LD and the imputed panels performed similarly, reaching values very close to the HD panels, except in the pacific oyster dataset, where the LD panel performed better than the imputed panel. In the fish species, when LD panels were used for genomic prediction without imputation, selection of markers based on either physical or genetic distance (instead of randomly) resulted in a high prediction accuracy, whereas imputation achieved near maximal prediction accuracy independently of the LD panel, showing higher reliability. Our results suggests that, in fish species, well-selected LD panels may achieve near maximal genomic selection prediction accuracy, and that the addition of imputation will result in maximal accuracy independently of the LD panel. These strategies represent effective and affordable methods to incorporate genomic selection into most aquaculture settings.

Assessment of Marine Gill Disease in Farmed Atlantic Salmon (Salmo salar) in Chile Using a Novel Total Gross Gill Scoring System: A Case Study.

Gill disorders have become more prevalent and widespread in finfish aquaculture in recent years. Their aetiology is often considered to be multifactorial. Effective diagnosis, control and prevention are hindered by the lack of standardised methodologies to characterise the aetiological agents, which produce an array of clinical and pathological presentations. The aim of this study was to define a novel gross pathological scoring system suitable for field-based macroscopic assessment of complex or multifactorial gill disease in farmed Atlantic salmon, using samples derived from a gill disease outbreak in Chile. Clinical assessment of gross gill morphology was performed, and gill samples were collected for qPCR and histology. A novel total gill scoring system was developed, which assesses gross pathological changes combining both the presumptive or healed amoebic gill disease (AGD) and the presence of other types of gill lesions. This scoring system offers a standardised approach to characterise the severe proliferative pathologies in affected gills. This total gill scoring system can substantially contribute to the development of robust mitigation strategies and could be used as an indicator trait for incorporating resistance to multifactorial gill disease into breeding goals.

Genomic Prediction Using Low Density Marker Panels in Aquaculture: Performance Across Species, Traits, and Genotyping Platforms.

Genomic selection increases the rate of genetic gain in breeding programs, which results in significant cumulative improvements in commercially important traits such as disease resistance. Genomic selection currently relies on collecting genome-wide genotype data accross a large number of individuals, which requires substantial economic investment. However, global aquaculture production predominantly occurs in small and medium sized enterprises for whom this technology can be prohibitively expensive. For genomic selection to benefit these aquaculture sectors, more cost-efficient genotyping is necessary. In this study the utility of low and medium density SNP panels (ranging from 100 to 9,000 SNPs) to accurately predict breeding values was tested and compared in four aquaculture datasets with different characteristics (species, genome size, genotyping platform, family number and size, total population size, and target trait). The traits show heritabilities between 0.19-0.49, and genomic prediction accuracies using the full density panel of 0.55-0.87. A consistent pattern of genomic prediction accuracy was observed across species with little or no accuracy reduction until SNP density was reduced below 1,000 SNPs (prediction accuracies of 0.44-0.75). Below this SNP density, heritability estimates and genomic prediction accuracies tended to be lower and more variable (93% of maximum accuracy achieved with 1,000 SNPs, 89% with 500 SNPs, and 70% with 100 SNPs). A notable drop in accuracy was observed between 200 SNP panels (0.44-0.75) and 100 SNP panels (0.39-0.66). Now that a multitude of studies have highlighted the benefits of genomic over pedigree-based prediction of breeding values in aquaculture species, the results of the current study highlight that these benefits can be achieved at lower SNP densities and at lower cost, raising the possibility of a broader application of genetic improvement in smaller and more fragmented aquaculture settings.

Optimizing Low-Cost Genotyping and Imputation Strategies for Genomic Selection in Atlantic Salmon.

Genomic selection enables cumulative genetic gains in key production traits such as disease resistance, playing an important role in the economic and environmental sustainability of aquaculture production. However, it requires genome-wide genetic marker data on large populations, which can be prohibitively expensive. Genotype imputation is a cost-effective method for obtaining high-density genotypes, but its value in aquaculture breeding programs which are characterized by large full-sibling families has yet to be fully assessed. The aim of this study was to optimize the use of low-density genotypes and evaluate genotype imputation strategies for cost-effective genomic prediction. Phenotypes and genotypes (78,362 SNPs) were obtained for 610 individuals from a Scottish Atlantic salmon breeding program population (Landcatch, UK) challenged with sea lice, Lepeophtheirus salmonis The genomic prediction accuracy of genomic selection was calculated using GBLUP approaches and compared across SNP panels of varying densities and composition, with and without imputation. Imputation was tested when parents were genotyped for the optimal SNP panel, and offspring were genotyped for a range of lower density imputation panels. Reducing SNP density had little impact on prediction accuracy until 5,000 SNPs, below which the accuracy dropped. Imputation accuracy increased with increasing imputation panel density. Genomic prediction accuracy when offspring were genotyped for just 200 SNPs, and parents for 5,000 SNPs, was 0.53. This accuracy was similar to the full high density and optimal density dataset, and markedly higher than using 200 SNPs without imputation. These results suggest that imputation from very low to medium density can be a cost-effective tool for genomic selection in Atlantic salmon breeding programs.

Enhancing genetic disease control by selecting for lower host infectivity and susceptibility.

Infectious diseases have a huge impact on animal health, production and welfare, and human health. Understanding the role of host genetics in disease spread is important for developing disease control strategies that efficiently reduce infection incidence and risk of epidemics. While heritable variation in disease susceptibility has been targeted in livestock breeding, emerging evidence suggests that there is additional genetic variation in host infectivity, but the potential benefits of including infectivity into selection schemes are currently unknown. A Susceptible-Infected-Recovered epidemiological model incorporating polygenic genetic variation in both susceptibility and infectivity was combined with quantitative genetics selection theory to assess the non-linear impact of genetic selection on field measures of epidemic risk and severity. Response to 20 generations of selection was calculated in large simulated populations, exploring schemes differing in accuracy and intensity. Assuming moderate genetic variation in both traits, 50% selection on susceptibility required seven generations to reduce the basic reproductive number R0 from 7.64 to the critical threshold of <1, below which epidemics die out. Adding infectivity in the selection objective accelerated the decline towards R0 < 1, to 3 generations. Our results show that although genetic selection on susceptibility reduces disease risk and prevalence, the additional gain from selection on infectivity accelerates disease eradication and reduces more efficiently the risk of new outbreaks, while it alleviates delays generated by unfavourable correlations. In conclusion, host infectivity was found to be an important trait to target in future genetic studies and breeding schemes, to help reducing the occurrence and impact of epidemics.

Can We Breed Cattle for Lower Bovine TB Infectivity?

Host resistance and infectivity are genetic traits affecting infectious disease transmission. This Perspective discusses the potential exploitation of genetic variation in cattle infectivity, in addition to resistance, to reduce the risk, and prevalence of bovine tuberculosis (bTB). In bTB, variability in M. bovis shedding has been previously reported in cattle and wildlife hosts (badgers and wild boars), but the observed differences were attributed to dose and route of infection, rather than host genetics. This article addresses the extent to which cattle infectivity may play a role in bTB transmission, and discusses the feasibility, and potential benefits from incorporating infectivity into breeding programmes. The underlying hypothesis is that bTB infectivity, like resistance, is partly controlled by genetics. Identifying and reducing the number of cattle with high genetic infectivity, could reduce further a major risk factor for herds exposed to bTB. We outline evidence in support of this hypothesis and describe methodologies for detecting and estimating genetic parameters for infectivity. Using genetic-epidemiological prediction models we discuss the potential benefits of selection for reduced infectivity and increased resistance in terms of practical field measures of epidemic risk and severity. Simulations predict that adding infectivity to the breeding programme could enhance and accelerate the reduction in breakdown risk compared to selection on resistance alone. Therefore, given the recent launch of genetic evaluations for bTB resistance and the UK government's goal to eradicate bTB, it is timely to consider the potential of integrating infectivity into breeding schemes.

Impact of Genetic Selection for Increased Cattle Resistance to Bovine Tuberculosis on Disease Transmission Dynamics.

Bovine tuberculosis (bTB) poses a challenge to animal health and welfare worldwide. Presence of genetic variation in host resistance to Mycobacterium bovis infection makes the trait amenable to improvement with genetic selection. Genetic evaluations for resistance to infection in dairy cattle are currently available in the United Kingdom (UK), enabling genetic selection of more resistant animals. However, the extent to which genetic selection could contribute to bTB eradication is unknown. The objective of this study was to quantify the impact of genetic selection for bTB resistance on cattle-to-cattle disease transmission dynamics and prevalence by developing a stochastic genetic epidemiological model. The model was used to implement genetic selection in a simulated cattle population. The model considered various levels of selection intensity over 20 generations assuming genetic heterogeneity in host resistance to infection. Our model attempted to represent the dairy cattle population structure and current bTB control strategies in the UK, and was informed by genetic and epidemiological parameters inferred from data collected from UK bTB infected dairy herds. The risk of a bTB breakdown was modeled as the percentage of herds where initially infected cows (index cases) generated secondary cases by infecting herd-mates. The model predicted that this risk would be reduced by half after 4, 6, 9, and 15 generations for selection intensities corresponding to genetic selection of the 10, 25, 50, and 70% most resistant sires, respectively. In herds undergoing bTB breakdowns, genetic selection reduced the severity of breakdowns over generations by reducing both the percentage of secondary cases and the duration over which new secondary cases were detected. Selection of the 10, 25, 50, and 70% most resistant sires reduced the percentage of secondary cases to <1% in 4, 5, 7, and 11 generations, respectively. Similarly, the proportion of long breakdowns (breakdowns in which secondary cases were detected for more than 365 days) was reduced by half in 2, 2, 3, and 4 generations, respectively. Collectively, results suggest that genetic selection could be a viable tool that can complement existing management and surveillance methods to control and ultimately eradicate bTB.

Quantitative genetic analysis of the bTB diagnostic single intradermal comparative cervical test (SICCT).

BACKGROUND: Bovine tuberculosis (bTB) is a disease of significant economic importance and is a persistent animal health problem with implications for public health worldwide. Control of bTB in the UK has relied on diagnosis through the single intradermal comparative cervical test (SICCT). However, limitations in the sensitivity of this test hinder successful eradication and the control of bTB remains a major challenge. Genetic selection for cattle that are more resistant to bTB infection can assist in bTB control. The aim of this study was to conduct a quantitative genetic analysis of SICCT measurements collected during bTB herd testing. Genetic selection for bTB resistance will be partially informed by SICCT-based diagnosis; therefore it is important to know whether, in addition to increasing bTB resistance, this might also alter genetically the epidemiological characteristics of SICCT. RESULTS: Our main findings are that: (1) the SICCT test is robust at the genetic level, since its hierarchy and comparative nature provide substantial protection against random genetic changes that arise from genetic drift and from correlated responses among its components due to either natural or artificial selection; (2) the comparative nature of SICCT provides effective control for initial skin thickness and age-dependent differences; and (3) continuous variation in SICCT is only lowly heritable and has a weak correlation with SICCT positivity among healthy animals which was not significantly different from zero (P > 0.05). These emerging results demonstrate that genetic selection for bTB resistance is unlikely to change the probability of correctly identifying non-infected animals, i.e. the test's specificity, while reducing the overall number of cases. CONCLUSIONS: This study cannot exclude all theoretical risks from selection on resistance to bTB infection but the role of SICCT in disease control is unlikely to be rapidly undermined, with any adverse correlated responses expected to be weak and slow, which allow them to be monitored and managed.

Genomic prediction for tuberculosis resistance in dairy cattle.

BACKGROUND: The increasing prevalence of bovine tuberculosis (bTB) in the UK and the limitations of the currently available diagnostic and control methods require the development of complementary approaches to assist in the sustainable control of the disease. One potential approach is the identification of animals that are genetically more resistant to bTB, to enable breeding of animals with enhanced resistance. This paper focuses on prediction of resistance to bTB. We explore estimation of direct genomic estimated breeding values (DGVs) for bTB resistance in UK dairy cattle, using dense SNP chip data, and test these genomic predictions for situations when disease phenotypes are not available on selection candidates. METHODOLOGY/PRINCIPAL FINDINGS: We estimated DGVs using genomic best linear unbiased prediction methodology, and assessed their predictive accuracies with a cross validation procedure and receiver operator characteristic (ROC) curves. Furthermore, these results were compared with theoretical expectations for prediction accuracy and area-under-the-ROC-curve (AUC). The dataset comprised 1151 Holstein-Friesian cows (bTB cases or controls). All individuals (592 cases and 559 controls) were genotyped for 727,252 loci (Illumina Bead Chip). The estimated observed heritability of bTB resistance was 0.23±0.06 (0.34 on the liability scale) and five-fold cross validation, replicated six times, provided a prediction accuracy of 0.33 (95% C.I.: 0.26, 0.40). ROC curves, and the resulting AUC, gave a probability of 0.58, averaged across six replicates, of correctly classifying cows as diseased or as healthy based on SNP chip genotype alone using these data. CONCLUSIONS/SIGNIFICANCE: These results provide a first step in the investigation of the potential feasibility of genomic selection for bTB resistance using SNP data. Specifically, they demonstrate that genomic selection is possible, even in populations with no pedigree data and on animals lacking bTB phenotypes. However, a larger training population will be required to improve prediction accuracies.