[Show abstract][Hide abstract] ABSTRACT: In the current study, principal component (PC) analysis was used to reduce the number of predictors in the estimation of direct genomic breeding values (DGV) for meat traits in a sample of 479 Italian Simmental bulls. SNP marker genotypes were determined with the 54K Illumina beadchip. After edits, 457 bulls and 40,179 SNPs were retained. PC extraction was carried out separately for each chromosome and 2,466 new variables able to explain 70% of total variance were obtained. Bulls were divided into reference and validation population. Three scenarios of the ratio reference:validation were tested: 70:30, 80:20, 90:10. Effect of PC scores on polygenic EBVs was estimated in the reference population using different models and methods. Traits analyzed were seven beef trait: daily live weight gain, size score, muscularity score, feet and legs score, beef index (economic index), calving ease direct effect, and cow muscularity. Accuracy was calculated as correlation between DGV and polygenic EBV in the validation bulls. Muscularity, feet and legs, and the beef index showed the highest accuracies calving ease the lowest. In general, accuracies were slightly higher when reference animals were selected at random and the best scenario was 90:10 and no substantial differences in accuracy were found among different methods. PC Analysis is entirely based on the factorisation of the SNP (co)variance matrix and produced a reduced set of variables (6% of the original variables) which may be used for different phenotypic traits. In spite of this huge reduction in the number of independent variables, DGV accuracies resulted similar to those obtained by using the whole set of SNP markers. Accuracies of direct genomic values found in the present work were always higher than those of traditional Parental average (PA). Thus, results of the present study may suggest a possible advantage of use of genomic indexes in the pre-selection of performance test candidates for beef traits. Moreover the relevant reduction of variable space might allow GS implementation also in small populations.
Journal of Animal Science 10/2012; · 2.09 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The large number of markers available compared with phenotypes represents one of the main issues in genomic selection. In this work, principal component analysis was used to reduce the number of predictors for calculating genomic breeding values (GEBV). Bulls of 2 cattle breeds farmed in Italy (634 Brown and 469 Simmental) were genotyped with the 54K Illumina beadchip (Illumina Inc., San Diego, CA). After data editing, 37,254 and 40,179 single nucleotide polymorphisms (SNP) were retained for Brown and Simmental, respectively. Principal component analysis carried out on the SNP genotype matrix extracted 2,257 and 3,596 new variables in the 2 breeds, respectively. Bulls were sorted by birth year to create reference and prediction populations. The effect of principal components on deregressed proofs in reference animals was estimated with a BLUP model. Results were compared with those obtained by using SNP genotypes as predictors with either the BLUP or Bayes_A method. Traits considered were milk, fat, and protein yields, fat and protein percentages, and somatic cell score. The GEBV were obtained for prediction population by blending direct genomic prediction and pedigree indexes. No substantial differences were observed in squared correlations between GEBV and EBV in prediction animals between the 3 methods in the 2 breeds. The principal component analysis method allowed for a reduction of about 90% in the number of independent variables when predicting direct genomic values, with a substantial decrease in calculation time and without loss of accuracy.
Journal of Dairy Science 06/2012; 95(6):3390-400. · 2.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: The aim of this study was to investigate the effects of CSN2-CSN3 (beta-kappa-casein) haplotypes and BLG (beta-lactoglobulin) genotypes on milk production traits, content of protein fractions, and detailed protein composition of individual milk of Simmental cows. Content of the major protein fractions was measured by reversed-phase HPLC in individual milk samples of 2,167 cows. Protein composition was measured as percentage of each casein (CN) fraction to total CN and as percentage of beta-lactoglobulin (beta-LG) to total whey protein. Genotypes at CSN2, CSN3, and BLG were ascertained by reversed-phase HPLC, and CSN2-CSN3 haplotype probabilities were estimated for each cow. Traits were analyzed by using a linear model including the fixed effects of herd-test-day, parity, days in milk, and somatic cell score class, linear regressions on haplotype probabilities, class of BLG genotype, and the random effect of the sire of the cow. Effects of haplotypes and BLG genotypes on yields were weak or trivial. Genotype BB at BLG and haplotypes carrying CSN2 B and CSN3 B were associated with increased CN content and CN number. Haplotypes including CSN3 B were associated with increased kappa-CN content and percentage of kappa-CN to total CN and with decreased percentages of alpha(S1)- and gamma-CN to total CN. Allele CSN2 B had the effect of increasing beta-CN content and decreasing content of alpha(S1)-CN. Haplotypes including allele CSN2 A(1) exhibited decreased beta-, alpha(S2)-, and gamma-CN concentrations and increased alpha(S1)- and kappa-CN contents, whereas CSN2 I had positive effects on beta-CN concentration and trivial effects on content of other protein fractions. Effects of haplotypes on CN composition were similar to those exerted on content of CN fractions. Allele BLG A was associated with increased beta-LG concentration and percentage of beta-LG to total whey protein and with decreased content of other milk proteins, namely beta-CN and alpha(S1)-CN. Estimated additive genetic variance for investigated traits ranged from 14 to 39% of total variance. Increasing the frequency of specific genotypes or haplotypes by selective breeding might be an effective way to change milk protein composition.
Journal of Dairy Science 08/2010; 93(8):3797-808. · 2.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Genetic relationships between lactation curve traits and Somatic Cell Count are of great interest for dairy cattle
breeding. Factor Analysis (MFA) and Principal Component Analysis (PCA) can be used to extract from the correlation
matrix of milk test day records new unobservable (latent) variables that can be related to lactation curve
shape. Previous researches report that MFA is particularly able to extract two latent variables related with level
of production in early lactation (PEL) and lactation persistency (PERS), respectively, whereas PCA yields a leading
component related to the average level of production (AVY) for the whole lactation and a second component negatively
related with tests of early lactation and positively with tests of the second part of lactation (SLOPE). Aim of
this work was to estimate genetic correlations between lactation curve shape traits and Somatic Cell Score (SCS).
MFA and PCA were carried out on a data set of 16,020 lactations of Italian Simmental cows, each with six TD
records for milk yield recorded with the A4 scheme. Genetic parameters were estimated with a bivariate animal
model that included fixed effects of herd-test date, parity*age*lactation stage (only parity*age for lactation curve
traits), calving season, and random effects of additive genetic and permanent environment. Heritability estimates
were moderate for lactation curve traits (0.15, 0.15, 0.21 and 0.09 for PEL, PERS, AVY and SLOPE, respectively)
and low for SCS (0.09). Correlations between lactation curve traits and SCS were favourable, i.e. negative, except
for the level of production in early lactation. In particular, the genetic improvement of lactation persistency result
in a contemporary reduction of SCS (rg -0.55 and -0.51 with PERS and SLOPE, respectively) whereas the increase
of level of production in early lactation can lead to a moderate increase of SCS (rg 0.13). Finally, the two measures
of persistency could be used for different selection strategies: the use of PERS may allow for the increase of persistency
together with the total lactation yield whereas the use of SLOPE may result in an improvement of the lactation
curve shape without modifying total lactation yield.
[Show abstract][Hide abstract] ABSTRACT: In Random Regression models (RRM), the most updated version of Test Day (TD) models, the lactation curve is split into a fixed average curve and a random animal specific part (deviation from the average curve) (Schaeffer, 2004). The variance component of the RR coefficients determines the (co) variance function of each pair of days in milk (DIM) (Pool and Meuwissen, 2000).
[Show abstract][Hide abstract] ABSTRACT: Estimated breeding values for Calving Ease for Italian Simmental bulls are not available, so it is not possible to develop appropriate mating schemes.The aim of this study was to estimate genetic parameters for calving ease in Italian Simmental cows, using a bivariate linear animal model. Calving ease scores for first and later parities cows were considered as different but correlated traits. The model accounted for contemporary groups, sex of calf, season of calving within region, and age at calving within parity effects; both direct and maternal genetic effects were considered. Heritabilities of direct (4.9-3.2%) and maternal (3.4%-1.2%) effects were comparable to values reported for other Simmental populations. The genetic correlation between first and later parities calving ease was 0.974 and 0.779 for direct and maternal genetic effects, respectively. The genetic correlation between direct and maternal effects was –0.331 and –0.412 for first and later parities, respectively. Genetic evaluation for calving difficulty is feasible, but is necessary to improve the quality of the data.
[Show abstract][Hide abstract] ABSTRACT: Several investigations have recently searched for significant association between gene polymorphisms
and milk traits in livestock and model species. In several cases, it remains rather difficult to assess if
the observed effects are caused by the mutation tested, by a nearby mutation in the same gene or by a
mutation in a different gene or DNA region in linkage disequilibrium with the former. As a consequence,
only in a few cases (e.g., κ-casein, SCD, DGAT1) the causative mutation seems to have been identified
and, even when evidence is rather clear, genetic heterogeneity and genetic background may influence the
size of allele substitution effects. Therefore, the significance of gene-trait associations and the estimate
of their effect have to be verified in any new population in which this information is planned to be used,
to estimate its actual utility in gene assisted breeding. In the SelMol project, we selected 29 candidate
genes on the basis of known relationships between physiological or biochemical processes and evidence of
significant association with milk traits in cattle, in related (e.g., sheep and goats) and model (e.g., mouse)
species. A total of 106 SNPs were selected, using either information available in literature, or in silico,
searching the NCBI dbSNP database. SNPs found significantly associated in other investigations were
preferentially targeted. Otherwise non-synonymous SNPs and those in putative control regions (e.g., in
promoter binding sites) were selected from dbSNP. If within a gene no SNP having one of these characteristics
was available in dbSNP, synonymous SNPs, occurring in introns and untranslated non-control
regions were chosen. DNA was extracted from semen of elite sires. SNPs polymorphism was confirmed by
screening a panel of 32 individuals each of Pezzata Rossa (PR), Bruna Italiana (BI), and Frisona Italiana
(FI) dairy cattle breeds. A total of 73 SNPs were confirmed as polymorphic in at least one breed: 63 in PR,
61 in BI, and 68 in FI. Polymorphic SNPs were genotyped on 400 individuals of PR and 600 of BI. Statistical
tests were applied to detect selection sweeps, significant association to EBVs and phenotypic traits
related to milk production and quality (milk yield, protein and fat yield and percentage), together with a
number of functional traits (fertility, SCS as indicator of mastitis resistance, conformational traits, and
[Show abstract][Hide abstract] ABSTRACT: Lactation persistency is defined as the ability of a cow to maintain as high as possible milk daily yield during lactation. This property, also so called "flatness" of lactation curve, has positive effects on health and reproduction status, because of a less negative energetic balance. Principal component analysis were used to decompose correlation matrix of test-day milk yields of 145,812 lactations of 90,655 Italian Simmental cows. The two leading principal components were associated, with whole lactation yield and lactation curve shape, respectively. The second principal component (PC2) was treated as new quantitative phenotype and analized under a BLUP animal model for a genetic evaluation. Genetic trends of cows and bulls show a substantial flat pattern. Empirical comparison among average lactation pattern of groups daughters of bulls which were very different on PC2 ebvs, confirmed the goodness of this indicator. Correlations between these ebvs and those calculated in Germany/Austria on bulls evaluated in both countries were moderate, that means ebvs for this trait calculated in this study are quite comparable.
[Show abstract][Hide abstract] ABSTRACT: Multivariate factor analysis and principal component analysis were used to decompose the correlation matrix of test-day milk yields of 48,374 lactations of 21,721 Italian Simmental cows. Two common latent factors related to level of production in early lactation and lactation persistency, and 2 principal components associated with the whole lactation yield and persistency were obtained. Factor and principal component scores were treated as new quantitative phenotypes related to prominent features of lactation curve shape. Genetic parameters were estimated by univariate and bivariate animal models. Estimates of heritability were moderately low for both latent factors (0.13 for persistency and yield early in lactation). Heritabilities of the principal component related to total lactation yield and 305-d yield were similar (0.19 and 0.20, respectively). Finally, heritability was quite low for the principal component related to lactation persistency (0.07). Repeatabilities between lactations were about 0.27 for both latent factors, around 0.4 for the first principal component and 305-d yield, and 0.11 for the second principal component. Moderate genetic correlation among common factors (0.26) and their high genetic correlation with total lactation yield (>0.60) suggest that selection can be used to change the shape of lactation curve as well as improve yield. Scores of the second principal component can be used to genetically improve persistency while maintaining constant total lactation yield.
Journal of Dairy Science 08/2006; 89(8):3188-94. · 2.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: In italian Simmental milking speed is a very important trait. Nowadays milking speed is taken into account in italian Simmental selection index (IDA) with an economic weight of 7.5%. Since early '90s this trait is recorded as farmer's score collected by type classifiers. Recently, in order to get more reliable bulls proofs, italian Simmental breed organization(ANAPRI) started to collect also measured data on progeny-test bulls daughters and on bull's mothers. Therefore, milk recording agencies have been asked to collect also Total Milking Time (TMT) during official milk recording test-days. Milk flow (kg/m) i.e. ratio between milk yield by TMT, was analyzed. A bi-trait Blup Animal Model analysis has been developed estimating genetic parameters. Heritabilities were .15 and .20 respectively for farmer's score and milk flow. Genetic correlation between these two traits was high (.83). Breeding values have been also estimated for entire population observing a slightly favourable genetic trend. Pretty good correlations have been also observed between proofs of bulls evaluated also in other countries (Germany-Austria-France).
[Show abstract][Hide abstract] ABSTRACT: The study of relationships between mathematical properties of functions used to model lactation curves is usually limited to the evaluation of the goodness of fit. Problems related to the existence of different lactation curve shapes are usually neglected or solved drastically by considering shapes markedly different from the standard as biologically atypical. A deeper investigation could yield useful indications for developing technical tools aimed at modifying the lactation curve in a desirable fashion. Relationships between mathematical properties and lactation curve shapes were analyzed by fitting several common functions (Wood incomplete gamma, Wilmink's exponential, Ali and Schaeffer's polynomial regression, and fifth-order Legendre polynomials) to 229,518 test-day records belonging to 27,837 lactations of Italian Simmental cows. Among the best fits (adjusted r(2) higher than 0.75), the 3-parameter models (Wood and Wilmink) were able to detect 2 main groups of curve shape: standard and atypical. Five-parameter models (Ali and Schaeffer function and the Legendre polynomials) were able to recognize a larger number of curve shapes. The higher flexibility of 5-parameter models was accompanied by increased sensitivity to local random variation as evidenced by the bias in estimated test-day yields at the beginning and end of lactation (border effect). Meaning of parameters, range of their values and of their (co) variances are clearly different among groups of curves. Our results suggest that analysis based on comparisons between parameter values and (co)variances should be done carefully. Comparisons among parameter values and (co)variances could yield more robust, reliable, and easy to interpret results if performed within groups based on curve shape.
Journal of Dairy Science 04/2005; 88(3):1178-91. · 2.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Milk test-day records of 5728 lactations of Italian
Simmental cows were analyzed with multivariate factor
analysis in order to extract 2 common factors, whose
scores were used as quantitative measures of 2 main
features of lactation curve shape—i.e., the increasing
rate of yield in the first part of lactation and the rate
of decline of milk yield after the lactation peak. The 2
indices, objectively derived from the correlation matrix
of original test-day records, showed a high discriminant
power in separating lactation curves with different
shapes. The weak correlation between the 2 factors
(0.11), together with the high correlation of factors and
the total 305-d yield (about 0.70), suggests that an increase
in lactation yield could be achieved by acting
only on one of the 2 factors related to lactation-curve
shape, with the other kept constant at a medium or low
value. The suitability of the 2 factors as descriptors of
lactation patterns has been confirmed by the relationships
found between factor scores and the main environmental
effects known to affect the shape of the lactation
curve, such as parity and season of calving.
Journal of Dairy Science 05/2004; · 2.57 Impact Factor
[Show abstract][Hide abstract] ABSTRACT: Autoregressive Moving Average (ARMA) models, originally developed in the contest of time series analysis, were used to predict Test Day (TD) yields of milk production traits in dairy cows. ARMA models areable to take into account both the average lactation curve of homogeneous groups of animals and the residual individual variability that may be explained in terms of probability models, such as Autoregressive (AR) and Moving Average (MA) processes. Milk, fat, and protein yields of 6000 Italian Simmental cows with 8 TD records per lactation were analyzed. Data were grouped according to parity (1st, 2nd, and 3rd calving) and fitted to a Box-Jenkins ARMA model in order to predict TD yields in five situations of incomplete lactations. Reasonable accuracies have been obtained for a limited horizon of prediction: average correlations among actual and predicted data were 0.85, 0.72, and 0.80 for milk, fat and protein yields when the first predicted TD was one step ahead (on average 42d) of the last actual record available. Cumulative 305-d yields were calculated using all actual (actual yields) or actual plus forecasted (estimated yields) daily yields. Accuracy of lactation predictions was remarkable even when only a few actual TD records were available, with values of 0.88 for milk and protein and 0.84 for fat for the correlations between actual and estimated yields when 6 out of 8 TD records were predicted. Accuracy rapidly increases with the number of actual TD available: correlations were about 0.96 for milk and protein and 0.93 for fat when 4 out of 8 TD records were predicted. In comparison with other prediction methods, ARMA modelsare very simple and can be easily implemented in data recording software, even at the farm level.
Journal of Dairy Science 12/2002; 85(11):3107-14. · 2.57 Impact Factor