Massimo Cellesi

Use of the canonical discriminant analysis to select SNP markers for bovine breed assignment and traceability purposes

Several market research studies have shown that consumers are primarily concerned with the provenance of the food they eat. Among the available identification methods, only DNA-based techniques appear able to completely prevent frauds. In this study, a new method to discriminate among different bovine breeds and assign new individuals to groups was developed. Bulls of three cattle breeds farmed in Italy - Holstein, Brown, and Simmental - were genotyped using the 50K SNP Illumina BeadChip. Multivariate canonical discriminant analysis was used to discriminate among breeds, and discriminant analysis (DA) was used to assign new observations. This method was able to completely identify the three groups at chromosome level. Moreover, a genome-wide analysis developed using 340 linearly independent SNPs yielded a significant separation among groups. Using the reduced set of markers, the DA was able to assign 30 independent individuals to the proper breed. Finally, a set of 48 high discriminant SNPs was selected and used to develop a new run of the analysis. Again, the procedure was able to significantly identify the three breeds and to correctly assign new observations. These results suggest that an assay with the selected 48 SNP could be used to routinely track monobreed products.

Detection of selection signatures in Piemontese and Marchigiana cattle, two breeds with similar production aptitudes but different selection histories

BackgroundDomestication and selection are processes that alter the pattern of within- and between-population genetic variability. They can be investigated at the genomic level by tracing the so-called selection signatures. Recently, sequence polymorphisms at the genome-wide level have been investigated in a wide range of animals. A common approach to detect selection signatures is to compare breeds that have been selected for different breeding goals (i.e. dairy and beef cattle). However, genetic variations in different breeds with similar production aptitudes and similar phenotypes can be related to differences in their selection history.MethodsIn this study, we investigated selection signatures between two Italian beef cattle breeds, Piemontese and Marchigiana, using genotyping data that was obtained with the Illumina BovineSNP50 BeadChip. The comparison was based on the fixation index (Fst), combined with a locally weighted scatterplot smoothing (LOWESS) regression and a control chart approach. In addition, analyses of Fst were carried out to confirm candidate genes. In particular, data were processed using the varLD method, which compares the regional variation of linkage disequilibrium between populations.ResultsGenome scans confirmed the presence of selective sweeps in the genomic regions that harbour candidate genes that are known to affect productive traits in cattle such as DGAT1, ABCG2, CAPN3, MSTN and FTO. In addition, several new putative candidate genes (for example ALAS1, ABCB8, ACADS and SOD1) were detected.ConclusionsThis study provided evidence on the different selection histories of two cattle breeds and the usefulness of genomic scans to detect selective sweeps even in cattle breeds that are bred for similar production aptitudes.

Dissection of genomic correlation matrices of US Holsteins using multivariate factor analysis

The aim of this study was to compare correlation matrices between direct genomic predictions for 31 traits at the genomic and chromosomal levels in US Holstein bulls. Multivariate factor analysis carried out at the genome level identified seven factors associated with conformation, longevity, yield, feet and legs, fat and protein content traits. Some differences were found at the chromosome level; variations in covariance structure on BTA 6, 14, 18 and 20 were interpreted as evidence of segregating QTL for different groups of traits. For example, milk yield and composition tended to join in a single factor on BTA 14, which is known to harbour the DGAT1 locus that affects these traits. Another example was on BTA 18, where a factor strongly correlated with sire calving ease and conformation traits was identified. It is known that in US Holstein, there is a segregating QTL on BTA18 influencing these traits. Moreover, a possible candidate gene for daughter pregnancy rate was suggested for BTA28. The methodology proposed in this study could be used to identify individual chromosomes, which have covariance structures that differ from the overall (whole genome) covariance structure. Such differences can be difficult to detect when a large number of traits are evaluated, and covariances may be affected by QTL that do not have large allele substitution effects.

Use of partial least squares regression to impute SNP genotypes in Italian cattle breeds.

BackgroundThe objective of the present study was to test the ability of the partial least squares regression technique to impute genotypes from low density single nucleotide polymorphisms (SNP) panels i.e. 3K or 7K to a high density panel with 50K SNP. No pedigree information was used.MethodsData consisted of 2093 Holstein, 749 Brown Swiss and 479 Simmental bulls genotyped with the Illumina 50K Beadchip. First, a single-breed approach was applied by using only data from Holstein animals. Then, to enlarge the training population, data from the three breeds were combined and a multi-breed analysis was performed. Accuracies of genotypes imputed using the partial least squares regression method were compared with those obtained by using the Beagle software. The impact of genotype imputation on breeding value prediction was evaluated for milk yield, fat content and protein content.ResultsIn the single-breed approach, the accuracy of imputation using partial least squares regression was around 90 and 94% for the 3K and 7K platforms, respectively; corresponding accuracies obtained with Beagle were around 85% and 90%. Moreover, computing time required by the partial least squares regression method was on average around 10 times lower than computing time required by Beagle. Using the partial least squares regression method in the multi-breed resulted in lower imputation accuracies than using single-breed data. The impact of the SNP-genotype imputation on the accuracy of direct genomic breeding values was small. The correlation between estimates of genetic merit obtained by using imputed versus actual genotypes was around 0.96 for the 7K chip.ConclusionsResults of the present work suggested that the partial least squares regression imputation method could be useful to impute SNP genotypes when pedigree information is not available.

Genome-wide association analysis in Italian Simmental cows for lactation curve traits using a low-density (7K) SNP panel

High-throughput cow genotyping has opened new perspectives for genome-wide association studies (GWAS). Directly recorded phenotypes and several records per animal could be used. In this study, a GWAS on lactation curve traits of 337 Italian Simmental cows genotyped with the Illumina (San Diego, CA) low-density BeadChip (7K) was carried out. Scores of the first 2 principal components extracted from test-day records (7 for each lactation) for milk yield, fat and protein percentages, and somatic cell score were used as phenotypes. The first component described the average level of the lactation curve, whereas the second summarized its shape. Data were analyzed with a mixed linear model that included fixed effects of herd, calving month, calving year, parity, SNP genotype, and random effects of animal and permanent environment. All statistically significant markers (Bonferroni corrected) were associated with the average level component (2 for milk yield, 9 for fat percentage, 6 for protein percentages, and 1 for somatic cell score). No markers were found to be associated with the lactation curve shape. Gene discovery was performed using windows of variable size, according to the linkage disequilibrium level of the specific genomic region. Several suggestive candidate genes were identified, some of which already reported to be associated with dairy traits, such as DGAT1. Others were involved in lipid metabolism, in protein synthesis, in the immune response, in cellular processes, and in early development. The large number of genes flagged in the present study suggests interesting perspectives for the use of low-density genotyped females for GWAS, also for novel phenotypes that are not currently considered as breeding goals.

Multiple-breed genomic evaluation by principal component analysis in small size populations

In this study, the effects of breed composition and predictor dimensionality on the accuracy of direct genomic values (DGV) in a multiple breed (MB) cattle population were investigated. A total of 3559 bulls of three breeds were genotyped at 54 001 single nucleotide polymorphisms: 2093 Holstein (H), 749 Brown Swiss (B) and 717 Simmental (S). DGV were calculated using a principal component (PC) approach for either single (SB) or MB scenarios. Moreover, DGV were computed using all SNP genotypes simultaneously with SNPBLUP model as comparison. A total of seven data sets were used: three with a SB each, three with different pairs of breeds (HB, HS and BS), and one with all the three breeds together (HBS), respectively. Editing was performed separately for each scenario. Reference populations differed in breed composition, whereas the validation bulls were the same for all scenarios. The number of SNPs retained after data editing ranged from 36 521 to 41 360. PCs were extracted from actual genotypes. The total number of retained PCs ranged from 4029 to 7284 in Brown Swiss and HBS respectively, reducing the number of predictors by about 85% (from 82% to 89%). In all, three traits were considered: milk, fat and protein yield. Correlations between deregressed proofs and DGV were used to assess prediction accuracy in validation animals. In the SB scenarios, average DGV accuracy did not substantially change when either SNPBLUP or PC were used. Improvement of DGV accuracy were observed for some traits in Brown Swiss, only when MB reference populations and PC approach were used instead of SB-SNPBLUP (+10% HBS, +16%HB for milk yield and +3% HBS and +7% HB for protein yield, respectively). With the exclusion of the abovementioned cases, similar accuracies were observed using MB reference population, under the PC or SNPBLUP models. Random variation owing to sampling effect or size and composition of the reference population may explain the difficulty in finding a defined pattern in the results.

Maximum difference analysis: a new empirical method for genome-wide association studies

Abstract The availability of high-density single nucleotide polymorphism (SNPs) panels for humans and, recently, for several livestock species has given a great impulse to genome-wide association studies towards the identification of genes associated with complex traits and diseases. The frequentist and the Bayesian approach are commonly used to investigate marker associations with traits of interest. Briefly, the former is the most widely used method, being intuitive and easily to apply, whereas the latter requires deeper statistical knowledge, but has the advantage to include prior information to obtain a posterior probability of association. Both methods, however, require parameters or distributions to be set a priori by the researcher. In this work, we suggest a new empirical method for genome-wide studies (GWAS), which verifies marker-trait associations using the bootstrap resampling and Chebyshev’s inequality. This method, called Maximum Difference Analysis (MDA), was tested on a real dataset of 2093 Italian Holstein bulls with the objective of finding associations between SNPs and milk, fat and protein yield and fat and protein percentage. Results of the MDA method were compared with those obtained to a genome-wide association analysis performed using the R package GenABEL. In addition, we assessed the bovine annotated genes related to the traits under study. The MDA method was able to locate known important loci for milk productive traits, such as the DGAT1, PRLR, GHR and SCD. Moreover, some new putative candidate genes were detected. The python script of MDA procedure is available at www.animalbreeding.uniss.it.