Multivariate Restricted Maximum Likelihood Research Articles

Multivariate restricted maximum likelihood estimation of genetic parameters for growth, carcass and meat quality traits in French Large White and French Landrace pigs

ADG2, DP, ECLC and MQI were respectively 0.30, 0.64, 0.22, 0.52, 0.39, 0.60, 0.33 in the LW and 0.34, 0.56, 0.25, 0.46, 0.31, 0.68, 0.23 in the LR breed. Common litter effects ranged from 5% (ABT in LW breed) to 16% (ADG2 in LR breed) of phenotypic variance. Growth traits and FCR exhibited favourable genetic correlations, but were unfavourably correlated to DP and carcass lean content. MQI also showed unfavourable though generally low genetic correlations with all the other traits. These antagonisms were apparent in both breeds, but tended to be larger in the LW than in the LR breed.

Genetic parameters of test day records of British Holstein-Friesian heifers

AbstractEstimates of genetic parameters for test day records of yields of milk, fat and protein and concentrations of fat and protein were obtained on 47 736 British Holstein-Friesian heifers in 7973 herds, progeny of 40 proven (to improve connectedness) and 707 young sires (comprising about one-fifth of the progeny), using multivariate restricted maximum likelihood methods with a sire model.Heritability estimates for lactation yields of milk, fat and protein and concentrations of fat and protein were 0·49, 0·39, 0·43, 0·63 and 0·47, respectively. Estimates for individual test day records of these traits ranged from 0·27 to 0·43, 0·16 to 0·34, 0·22 to 0·33, 0·11 to 0·48 and 0·21 to 0·43, respectively. Generally, heritability estimates for test day records were lowest at start and highest in mid lactation.Estimates of genetic correlations among yields of a trait on different test days ranged from 0·57 to 0·99, and for fat and protein concentrations from 0·34 to 0·99, the correlations being highest for adjacent tests. Phenotypic correlations were lower than genetic correlations. Genetic correlations of test day records with corresponding lactation traits were high (0·76 to 0·99), being highest in mid lactation.Genetic correlations of test day milk yield with test day yields and concentrations of fat and protein throughout the lactation were similar to those for complete lactation.The high heritabilities of test day yields and their high genetic correlations with complete lactation, except for the first 1 or 2 test days, suggest that lactation performance may be predicted from test days in early and mid lactation.

Genetic factors contributing to variation in littersize in British Large White gilts

The number of ova released (ovulation rate) by 516 Large White gilts born between 1986 and 1989 was recorded. The weight of the gilt at birth, weaning and time of ovulation rate measurement and her number of teats were also recorded. Parrowing data (number born alive and litter weight at birth) corresponding to the ovulation rate were recorded from 382 of the gilts, enabling calculation of prenatal survival (number born alive/ovulation rate). The data were analysed using univariate and multivariate restricted maximum likelihood (REML) techniques with an individual animal model. The additive genetic direct and maternal components of variance and the common family and residual environmental components of variance and the additive genetic and residual environmental covariances between traits were estimated. The univariate REML analyses showed that the additive genetic direct component was a significant source of variation for gilt weight at birth and weaning, teat number, ovulation rate on the left hand side, total ovulation rate and litter weight at birth. Common family environmental effects were significant sources of variation for gilt weights and teat number. The multivariate REML analyses indicated that the genetic correlations between total ovulation rate and ovulation rate from the left and right ovaries were close to unity, with an estimate of the heritability of total ovulation rate of 0.37±0.09. In the data from gilts that farrowed, the heritabilities of ovulation rate, number born alive and prenatal survival were 0.30±0.10, 0.09±0.06 and 0.00±0.00, respectively. The genetic correlation between ovulation rate and litter size was close to unity, suggesting that genetic variation in ovulation rate explains virtually all of the genetic variation in number born alive in the population of Large White gilts understudy.

Heritability of and genetic correlations among faecal egg counts and productivity traits in Romney sheep

Abstract A total of 2611 spring-born Romney ewe lambs, the progeny of 60 sires over 3 years, were studied on a Hawke's Bay farm to estimate the heritability of faecal egg count (FEC), and correlations between FEC and production traits. Weaning weight, an autumn weight, the weight gain between these 2 times, and yearling fleece weight were recorded in each of 3 years. A dag score was obtained in the final 2 years. An anthelmintic drench was administered to 1987-born lambs in mid January after which further treatment was withheld until April. In the 2 subsequent years, drench treatment was administered in the first week of December and further treatment withheld until the mean FEC in a small monitor group reached a target value of 1000-1500 eggs/g. Once the target value was reached, a faecal sample was obtained from all animals for an FEC, the dag score and autumn liveweight were taken, and routine drenching was then resumed. Variance components were estimated using a multivariate restricted maximum likelihood (REML) procedure, with a sire model. Heritabilities were low for the liveweights and gain traits (0.08 ± 0.04 to 0.13 ± 0.04), and moderate for fleece weight (0.41 ± 0.09), dag score (0.24 ± 0.08), and both FEC and loge (FEC + 1) (0.34 ± 0.08 and 0.27 ± 0.07 respectively). Phenotypic correlations between loge (FEC + 1) and all production traits were negative, but very low (−0.01 to −0.05). The corresponding genetic correlations were also negative and ranged from −0.05 to −0.36. Broadly similar correlations occurred between productivity traits and untransformed FEC. Phenotypic and genetic correlations between FEC or loge (FEC + 1) and dag score were positive, and thus favourable. Overall, the results demonstrate that breeding for low nematode egg output in lambs (and thus by implication improved resistance to nematode establishment) is feasible in a commercial ram breeding enterprise. However, to achieve any substantial improvement in growth rates or dag scores of lambs under untreated exposure to worm infection, an independent culling approach incorporating these traits is probably the simplest method at present.

Phenotypic and genetic variation in body weight, food intake and energy utilisation in Hereford cattle I. Performance test results

From body weight, food intake and carcass composition data on 542 Hereford bull calves, measuredfrom 200 to 400 days, several traits relating to the efficiency of beef cattle production were derived and analysed. Traits included body weight at various ages, weight gain, predicted carcass lean content, lean growth rate, food intake, food conversion ratio, lean food conversion ratio, food intake in relation to metabolic body weight, energy required for protein and fat deposition, and predicted maintenance expenditure. Maintenance expenditure and the costs of fat and protein deposition were calculated by two means,firstly from allometric equations describing fat and protein accretion, and secondly from a multiple regression of food intake on weight gain and predicted carcass lean content. The two methods gave different mean values, but the correlations between traits calculated by the two methods were almost all 1.00. Exponents for metabolic body weight derived from the two methods were 0.738 and 0.758, respectively. Genetic parameters were calculated using multivariate Restricted Maximum Likelihood techniques.Body weight, carcass composition and traits combining these measurements were moderately to strongly inherited whereas traits related to food intake and efficiency were weakly to moderately inherited. Energy used to deposit fat and lean was more strongly inherited than predicted maintenance expenditure, and these traits were genetically almost uncorrelated. Maintenance energy expenditure showed no genetic relationship with predicted carcass lean content. Efficiency and predicted maintenance expenditure were favourably correlated.

Dairy herd life in relation to linear type traits and production 1. Phenotypic and genetic analyses in pedigree type classified herds

AbstractThe relation between survival to complete lactations 2, 3 and 4 and both linear type and production traits of pedigree Holstein-Friesian dairy cattle in the United Kingdom was analysed. There were records on 23 071 pedigree animals which were themselves type classified as heifers. The linear and quadratic phenotypic regressions of survival on most type traits were significant, even with yield fitted as a covariate. Phenotypic regressions of survival on milk, fat and protein yield were significantly positive, but that on protein content significantly negative. Phenotypic correlations with survival were under 0·1 for the linear traits, 0·16 for a subjective total type score, and 0·14 for milk yield. Genetic correlations between survival and type traits, estimated by multivariate restricted maximum likelihood, were as high as 0·4 for several traits of the udder and teats and for total score, and higher than for milk yield or composition. Regressions of survival on estimated sire transmitting abilities gave rather lower estimates of genetic correlations, except for total score. The results indicate that in these pedigree and classifying herds the total score is a culling criterion in its own right.

Correlations Between Production Traits in First Lactation Holstein Cows

Genetic and phenotypic correlations between milk yield, fat (yield and percent, protein (yield and percent), and somatic cell count in first lactation Holstein cows were estimated using a multivariate restricted maximum likelihood algorithm. There were 18,189 daughters of 257 sires in 928 herds. Genetic correlations between pairs of yield traits were all positive (.73 to .88), but phenotypic correlations with somatic cell count were small and negative. Genetic correlations between somatic cell count, and fat percent, and protein percent were negative, −.11. Milk yield, fat yield, and protein yield had heritabilities of .36, .38, and .25.

Genetic and Phenotypic Statistics of Lactation Cell Counts in Different Lactations of Holstein Cows

Between February 1977 and February 1982, 680,246 monthly test-day observations of somatic cell count were taken for Holstein cows having completed 79,124 lactations in 941 herds on the Quebec Dairy Herd Analysis Service. Data were transformed to natural log scale, and analyses were separate within five parity groups. Two lactational measures of cell count, geometric mean, and weighted (by test-day milk yield) mean of the log of monthly cell counts were calculated for each lactation. Maximum likelihood, iterative Minimum Norm Quadratic Unbiased Estimation, and multivariate Restricted Maximum Likelihood procedures were used for estimation of genetic and phenotypic parameters.Repeatabilities of log test-day cell counts and log of test-day cell counts corrected for milk yield were the same at each parity, .36, .41, .42, .42, and .42 in first, second, third, fourth, and fifth and later parities. Repeatabilities of lactational measures of cell count were .33 between first and second parities, .40 between second and third parities, .13 between first and third parities, and .27 between first, second, and third parities.Heritabilities of measures of lactational cell count were small from .06 to .14 in the five parities. Genetic correlations between measures of lactational cell count in different parities were close to unity, .90 to .97.

Genetic parameters for dairy production of Australian Black and White cows

An initial study on the inheritance of first and second lactation dairy production in Australian Black and White cows is presented for the states of New South Wales, Victoria, Queensland and Tasmania. A total of 116,043 first and 14,767 second lactation records for 326 young and 415 old, and 200 young and 31 old sires, respectively, were analyzed using a multivariate Restricted Maximum Likelihood procedure. Analyses were carried out for a mixed model fitting herd-year-seasons and proven sires as fixed, young sires as random effects and age at calving as a linear and quadratic covariable. In addition, the effects of lactation length and month of calving were investigated. The within-HYS phenotypic variation was high with distinct differences between states. As a consequence, heritability estimates were low compared to literature values. Pooled estimates were 0.17, 0.15, 0.38, 0.13 and 0.25 for first lactation milk yield, fat yield, fat content, protein yield and protein content, respectively. Correlations among yields were high, ranging from 0.71 to 0.91 genetically, and from 0.90 to 0.95 phenotypically. Milk yield was negatively related to concentration of constituents, genetic and phenotypic correlations being of the order of −0.4 and −0.2, respectively. Heritabilities for second lactation milk yield, fat yield and fat content were 0.18, 0.17 and 0.45, respectively, of similiar magnitude to first lactation values. The respective genetic correlations between first and second lactation were 0.96, 0.98 and 0.98, suggesting that production in both lactations is genetically identical. These high estimates can be attributed to the method of analysis, which accounts for culling based on first lactation performance.

Estimates of genetic parameters for milk and fat yield for the first three lactations in British Friesian cows

ABSTRACTA multivariate restricted maximum-likelihood procedure was used to estimate variance and covariance components between and within sires. This method, which considered all lactations simultaneously, accounted for the bias in later lactation records due to selection on dairy performance. Analysis was carried out for a mixed model with herd-year-seasons as fixed and sires as random effects, and fitting lactation length, calving age and month within season of calving as covariables. The data included 26 176 first, 19 978 second and 14 868 third lactation records for 679 test sires, and were analysed in 13 subsets. Additional records for proven sires, treated as fixed effects, were included to improve the data structure.Estimates for all components were higher in later than in earlier data sets, probably, to a large extent, because of a scale effect. Pooled estimates of heritabilities for lactations 1 to 3 were 0·28, 0·19 and 0·24 for milk yield, and 0·27, 0·21 and 0·25 for fat yield. Genetic correlations of 0·91, 0·91 and 0·95 for milk yield, and 0·91, 0·91 and 0·99 for fat yield, were found between lactations 1 and 2, 1 and 3, and 2 and 3, respectively. Results suggest that dairy performance in all lactations is almost identical genetically.