Background In commercial pig production, reduction of harmful social behavioural traits, such as ear manipulation and tail biting, is of major interest. Moreover, farmers prefer animals that are easy to handle. The aim of this experiment was to determine whether selection on social breeding values (SBV) for growth rate in purebred pigs affects behaviour in a weighing crate, lesions from ear manipulation, and tail biting of their crossbred progeny. Data were collected on crossbred F1 pigs allocated to 274 pens, which were progeny of purebred Landrace sows and Yorkshire boars from a DanBred nucleus herd. Results Behaviour in the weighing crate scored on a three-level scale showed that groups of pigs with high SBV for growth rate were significantly calmer than groups of pigs with low SBV (P < 0.027). When the mean SBV in the group increased by 1 unit, the proportion of pigs that obtained a calmer score level was increased by 14%. A significant (p = 0.04), favourable effect of SBV was found on both the number of pigs with ear lesions in the group and the mean number of ear lesions per pig. For a 1 unit increase in mean SBV, the mean number of lesions per pig decreased by 0.06 from a mean of 0.98. Individual severity of ear lesions conditional upon the number of ear lesions was also significantly affected (p = 0.05) by the mean SBV in the group. In groups for which the mean SBV increased by 1 unit, the proportion of pigs that were observed with a lower severity score was increased by 20% on a three-level scale. Most pigs received no tail biting injuries and no effect of SBV was observed on the tail injury score. Conclusions After 7 weeks in the finisher unit, crossbred progeny with high SBV were calmer in the weighing crate and had fewer ear lesions. These results indicate that selection of purebred parents for SBV for growth rate will increase welfare in their crossbred progeny by decreasing the number of ear lesions and making them easier to handle.

Background In breeding programs, recording large-scale feed intake (FI) data routinely at the individual level is costly and difficult compared with other production traits. An alternative approach could be to record FI at the group level since animals such as pigs are normally housed in groups and fed by a shared feeder. However, to date there have been few investigations about the difference between group- and individual-level FI recorded in different environments. We hypothesized that group- and individual-level FI are genetically correlated but different traits. This study, based on the experiment undertaken in purebred DanBred Landrace (L) boars, was set out to estimate the genetic variances and correlations between group- and individual-level FI using a bivariate random regression model, and to examine to what extent prediction accuracy can be improved by adding information of individual-level FI to group-level FI for animals recorded in groups. For both bivariate and univariate models, single-step genomic best linear unbiased prediction (ssGBLUP) and pedigree-based BLUP (PBLUP) were implemented and compared. Results The variance components from group-level records and from individual-level records were similar. Heritabilities estimated from group-level FI were lower than those from individual-level FI over the test period. The estimated genetic correlations between group- and individual-level FI based on each test day were on average equal to 0.32 (SD = 0.07), and the estimated genetic correlation for the whole test period was equal to 0.23. Our results demonstrate that by adding information from individual-level FI records to group-level FI records, prediction accuracy increased by 0.018 and 0.032 compared with using group-level FI records only (bivariate vs. univariate model) for PBLUP and ssGBLUP, respectively. Conclusions Based on the current dataset, our findings support the hypothesis that group- and individual-level FI are different traits. Thus, the differences in FI traits under these two feeding systems need to be taken into consideration in pig breeding programs. Overall, adding information from individual records can improve prediction accuracy for animals with group records.

Background Average daily gain (ADG) in pigs is affected by both direct and social genetic effects (SGE). However, selection for SGE in purebreds has not conclusively been shown to improve ADG in crossbreds, and it is unknown whether SGE in purebreds are equal to those in crossbreds. Moreover, SGE may reflect dominance related behaviour, which is affected by the variation in body weight within a group. Therefore, we hypothesized that (a) there is a positive effect of parent average SGE estimated in purebred pigs on phenotypic ADG in crossbred offspring, and (b) there is an interaction between SGE on ADG and standard deviation in starting weight of pigs within the group. We also hypothesized that (c) social genetic variance for ADG exists in crossbred pigs, and (d) there is a favourable genetic correlation between SGE on ADG in purebred and crossbred pigs. Results We found a statistically significant interaction between the standard deviation in starting weight and SGE within groups, and conditioning on the mean standard deviation in starting weight, we found a favourable regression coefficient (0.37 ± 0.21) of ADG in crossbreds on SGE in purebreds. Variances for SGE were small in both Landrace (L) and Yorkshire (Y), and higher for SGE in both the dam and sire component of crossbred YL. The genetic correlations between SGE in purebreds and the dam or sire component of SGE in crossbreds were also favourable (0.52 ± 0.48 and 0.34 ± 0.42, respectively), although not significantly different from 0. Conclusions We confirmed that there is a positive effect of SGE estimated using purebred information on phenotypic ADG in crossbreds, and that the largest effect is achieved when the within-group variation in starting weight is small. Our results indicate that social genetic variance in crossbreds exists and that there is a favourable genetic correlation between social genetic effects in purebreds and crossbreds. Collectively, our results indicate that selection for SGE on ADG in purebreds in a nucleus farm environment with little competition for resources can improve ADG in crossbreds in a commercial environment.

BackgroundSocial genetic effects (SGE) are the effects of the genotype of one animal on the phenotypes of other animals within a social group. Because SGE contribute to variation in economically important traits for pigs, the inclusion of SGE in statistical models could increase responses to selection (RS) in breeding programs. In such models, increasing the relatedness of members within groups further increases RS when using pedigree-based relationships; however, this has not been demonstrated with genomic-based relationships or with a constraint on inbreeding. In this study, we compared the use of statistical models with and without SGE and compared groups composed at random versus groups composed of families in genomic selection breeding programs with a constraint on the rate of inbreeding.ResultsWhen SGE were of a moderate magnitude, inclusion of SGE in the statistical model substantially increased RS when SGE were considered for selection. However, when SGE were included in the model but not considered for selection, the increase in RS and in accuracy of predicted direct genetic effects (DGE) depended on the correlation between SGE and DGE. When SGE were of a low magnitude, inclusion of SGE in the model did not increase RS, probably because of the poor separation of effects and convergence issues of the algorithms. Compared to a random group composition design, groups composed of families led to higher RS. The difference in RS between the two group compositions was slightly reduced when using genomic-based compared to pedigree-based relationships.ConclusionsThe use of a statistical model that includes SGE can substantially improve response to selection at a fixed rate of inbreeding, because it allows the heritable variation from SGE to be accounted for and capitalized on. Compared to having random groups, family groups result in greater response to selection in the presence of SGE but the advantage of using family groups decreases when genomic-based relationships are used.

Background Physical removal of individuals from groups causes reductions in group sizes and changes in group composition, which may affect the predictive ability of estimates of indirect genetic effects of animals on phenotypes of group mates. We hypothesized that including indirect genetic effects of culled animals and of animals without phenotypes in the analysis affects estimates of genetic parameters, improves predictive ability, and reduces bias of predicted breeding values. We tested this by applying different editing procedures, i.e. omission of individuals or groups from the data, and genetic models, i.e. a classical and an indirect genetic model (IGM) without or with weighting of indirect genetic effects based on the relative proportion of time spent in the pen or space allowance. Data consisted of average daily gain for 123,567 pigs in 11,111 groups, from which 3% of individuals in 25% of groups were prematurely removed from the group. Results The estimate of total heritability was higher (0.29 to 0.34) than that of direct heritability (0.23 to 0.25) regardless of the editing procedures and IGM used. Omission of individuals or groups from the data reduced the predictive ability of estimates of indirect genetic effects by 8 to 46%, and the predictive ability of estimates of the combined direct and indirect genetic effects by up to 4%. Omission of full groups introduced bias in predicted breeding values. Weighting of indirect genetic effects reduced the predictive ability of their estimates by at least 19% and of the estimates of the combined direct and indirect genetic effects by 1%. Conclusions We identified significant indirect genetic effects for growth in pigs. Culled animals should neither be removed from the data nor accounted for by weighting their indirect genetic effects in the model based on the relative proportion of time spent in the pen or space allowance, because it will reduce predictive ability and increase bias of predicted breeding values. Information on culled animals is important for prediction of indirect genetic effects and must be accounted for in IGM analyses by including fixed regressions based on relative time spent within the pen or relative space allowance.

Abstract Background Several studies have found that the growth rate of a pig is influenced by the genetics of the group members (indirect genetic effects). Accounting for these indirect genetic effects in a selection program may increase genetic progress for growth rate. However, indirect genetic effects are small and difficult to predict accurately. Genomic information may increase the ability to predict indirect genetic effects. Thus, the objective of this study was to test whether including indirect genetic effects in the animal model increases the predictive performance when genetic effects are predicted with genomic relationships. In total, 11,255 pigs were phenotyped for average daily gain between 30 and 94 kg, and 10,995 of these pigs were genotyped. Two relationship matrices were used: a numerator relationship matrix ( $${\mathbf{A}}$$ A ) and a combined pedigree and genomic relationship matrix ( $${\mathbf{H}}$$ H ); and two different animal models were used: an animal model with only direct genetic effects and an animal model with both direct and indirect genetic effects. The predictive performance of the models was defined as the Pearson correlation between corrected phenotypes and predicted genetic levels. The predicted genetic level of a pig was either its direct genetic effect or the sum of its direct genetic effect and the indirect genetic effects of its group members (total genetic effect). Results The highest predictive performance was achieved when total genetic effects were predicted with genomic information (21.2 vs. 14.7%). In general, the predictive performance was greater for total genetic effects than for direct genetic effects (0.1 to 0.5% greater; not statistically significant). Both types of genetic effects had greater predictive performance when they were predicted with $${\mathbf{H}}$$ H rather than $${\mathbf{A}}$$ A (5.9 to 6.3%). The difference between predictive performances of total genetic effects and direct genetic effects was smaller when $${\mathbf{H}}$$ H was used rather than $${\mathbf{A}}$$ A . Conclusions This study provides evidence that: (1) corrected phenotypes are better predicted with total genetic effects than with direct genetic effects only; (2) both direct genetic effects and indirect genetic effects are better predicted with $${\mathbf{H}}$$ H than $${\mathbf{A}}$$ A ; (3) using $${\mathbf{H}}$$ H rather than $${\mathbf{A}}$$ A primarily improves the predictive performance of direct genetic effects.

Average daily gain (ADG) in pigs is affected by the so-called social (or indirect) genetic effects (SGE). However, SGE may differ between sexes because boars grow faster than gilts and their social behaviours differ. We hypothesized that direct genetic effects (DGE) and SGE for ADG in pigs differ between boars and gilts and that accounting for these differences will improve the predictive ability of a social genetic effects model (SGM). Our data consisted of ADG from 30 to 94 kg for 32,212 uncastrated males (boars) and 48,252 gilts that were raised in sex-specific pens. Data were analyzed using a univariate model with sex as a fixed effect and a bivariate model with ADG in boars and gilts as separate traits using both a classical animal model (CM) and a SGM.

Background In settings with social interactions, the phenotype of an individual is affected by the direct genetic effect (DGE) of the individual itself and by indirect genetic effects (IGE) of its group mates. In the presence of IGE, heritable variance and response to selection depend on size of the interaction group (group size), which can be modelled via a ‘dilution’ parameter ( d ) that measures the magnitude of IGE as a function of group size. However, little is known about the estimability of d and the precision of its estimate. Our aim was to investigate how precisely d can be estimated and what determines this precision. Methods We simulated data with different group sizes and estimated d using a mixed model that included IGE and d . Schemes included various average group sizes (4, 6, and 8), variation in group size (coefficient of variation ( CV ) ranging from 0.125 to 1.010), and three values of d (0, 0.5, and 1). A design in which individuals were randomly allocated to groups was used for all schemes and a design with two families per group was used for some schemes. Parameters were estimated using restricted maximum likelihood (REML). Bias and precision of estimates were used to assess their statistical quality. Results The dilution parameter of IGE can be estimated for simulated data with variation in group size. For all schemes, the length of confidence intervals ranged from 0.114 to 0.927 for d , from 0.149 to 0.198 for variance of DGE, from 0.011 to 0.086 for variance of IGE, and from 0.310 to 0.557 for genetic correlation between DGE and IGE. To estimate d , schemes with groups composed of two families performed slightly better than schemes with randomly composed groups. Conclusions Dilution of IGE was estimable, and in general its estimation was more precise when CV of group size was larger. All estimated parameters were unbiased. Estimation of dilution of IGE allows the contribution of direct and indirect variance components to heritable variance to be quantified in relation to group size and, thus, it could improve prediction of the expected response to selection in environments with group sizes that differ from the average size.

Background: We tested the hypothesis that optimum-contribution selection (OCS) with restrictions imposed during optimisation realises most of the long-term genetic gain realised by OCS without restrictions.Methods: We used stochastic simulation to estimate long-term rates of genetic gain realised by breeding schemes that applied OCS without and with restrictions imposed during optimisation, where long-term refers to generations 23 to 25 (approximately). Six restrictions were imposed. Five of these removed solutions from the solution space. The sixth removed records of selection decisions made at earlier selection times. We also simulated a conventional breeding scheme with truncation selection as a reference point. Generations overlapped, selection was for a single trait, and the trait was observed for all selection candidates prior to selection.Results: OCS with restrictions realised 67 to 99% of the additional gain realised by OCS without restrictions, where additional gain was the difference in the long-term rates of genetic gain realised by OCS without restrictions and our reference point with truncation selection. The only exceptions were those restrictions that removed all solutions near the optimum solution from the solution space and the restriction that removed records of selection decisions made at earlier selection times. Imposing these restrictions realised only -12 to 46% of the additional gain.Conclusions: Most of the long-term genetic gain realised by OCS without restrictions can be realised by OCS with restrictions imposed during optimisation, provided the restrictions do not remove all solutions near the optimum from the solution space and do not remove records of earlier selection decisions. In breeding schemes where OCS cannot be applied optimally because of biological and logistical restrictions, OCS with restrictions provides a useful alternative. Not only does it realise most of the long-term genetic gain, OCS with restrictions enables OCS to be tailored to individual breeding schemes.

Dairy cow mortality has been steadily increasing during the last 2 decades in Denmark. This study aims to verify whether genetic mechanisms might be contributing to this increase. To do so, the records of 880,480 Holstein, 142,306 Jersey, and 85,206 Red Danish dairy cows calving from 1990 to 2006 were retrieved from the Danish Cattle register. Two causes of culling of cows were considered: death and slaughtering. Bivariate competing risk genetic models with a sire model structure were used to describe the death and the slaughtering rates simultaneously. The models included 2 random components: a sire random component with pedigree representing the sire genetic effects and a herd-year-season component. Moreover, the level of heterozygosity and the sire breed proportions were included in the models as covariates to account for potential nonadditive genetic effects due to the massive introduction of genetic material from other populations. The correlations between the sire components for death rate and slaughter rate were negative and small for the 3 populations, suggesting the existence of specific genetic mechanisms for each culling reason and common concurrent genetic mechanisms. In the Holstein population, the effects of the changes in the level of heterozygosity, breed composition, and the increasing genetic trend acted in the same direction, increasing the death rate in recent years. In the Jersey population, the effects of the level of heterozygosity and the breed proportion were small, and only the increasing genetic trend can be pointed as a genetic cause to the observed increase in the mortality rate. In the Red Danish population, neither the time-development pattern of the genetic trend nor the changes in the level of heterozygosity and breed composition could be causing the observed increase in the mortality; thus, nongenetic factors must be causing this negative development.

This study aimed to estimate genetic parameters for the mortality causes stillborn, weak at birth, starvation, crushing, and miscellaneous in crossbred piglets produced by crossbred dams. Data were collected in a single Danish commercial herd from October 2006 to July 2008 and consisted of 34,194 piglets (2,152 litters), which originated from 195 Danish Duroc sires and 955 crossbreds between Danish Landrace and Danish Yorkshire dams. Of the 34,194 piglets born, 11.5% were stillborn, 4.2% were crushed by the sow, 2.7% died due to starvation, 2.3% were weak at birth, and 2.2% died of miscellaneous causes before weaning. The first 4 mentioned causes were analyzed multivariately using a generalized linear mixed model with a probit link function, including the genetic effect of both sire and dam. Heritabilities based on the sire component ranged between 0.08 for stillborn and 0.21 for starvation whereas heritabilities based on the dam component ranged between 0.01 for miscellaneous and 0.24 for stillborn, indicating that reducing piglet mortality through genetic selection is possible. The expected observed responses to selection would, however, be low. The genetic correlations between mortality traits based on the sire component ranged from -0.05 between stillborn and starvation to 0.35 between stillborn and weak at birth whereas genetic correlations based on the dam component ranged from -0.11 between weak at birth and starvation to 0.76 between crushing and starvation. There seemed to be a favorable relationship between the 2 traits stillborn and weak at birth and between crushing and starvation, implying that care should be taken with correct recordings of mortality causes. The genetic correlation precision was rather low, and if nonadditive effects are not accounted for, there may be unexpected correlated responses among the different mortality causes in the crossbred mortalities.

This study aimed to investigate the presence of genetic variation in footpad dermatitis (FPD) and hock burns (HB) and the possibility to genetically select against these. A field trial including 10 commercial broiler lines (n = 102 to 265) was carried out at 2 Dutch farms. Footpad dermatitis and HB were subjectively scored at approximately 4, 5, and 7 wk on a scale from 0 through 5. Genetic parameters were estimated in 2 lines based on a larger data set. The overall agreement of repeated FPD and HB scores was high (0.66 to 0.86) and the scoring system was, therefore, considered reliable. Kendall's tau between left and right scores was lower than 1 (FPD: 0.73 and HB: 0.57), and both left and right FPD and HB must, therefore, be evaluated. High prevalences of FPD, but also HB, were achieved in the field trial, but lower prevalences may be sufficient for genetic evaluations and would be less detrimental to welfare. Genetic variation between and within lines was present for both FPD and HB as indicated by between-line differences and heritabilities, and selection against FPD and HB is, therefore, possible. It is important that selection is done against both FPD and HB, and such selection should not have a negative influence on the genetic improvement in BW. In contrast, continued selection for increased BW while ignoring FPD in the breeding goal is likely to lead to an increased propensity to develop FPD in broilers.

Genetic factors influencing the outcome of bovine ovum pick-up–in vitro production (OPU-IVP) and its relation to female fertility were investigated. For the first time, genetic parameters were estimated for the number of cumulus-oocyte complexes (Ncoc), quality of cumulus-oocyte complexes (Qcoc), number and proportion of cleaved embryos at Day 4 (NcleavD4, PcleavD4), and number and proportion of total and transferable embryos at Day 7 of culture (NembD7, PembD7 and NTembD7, PTembD7, respectively). Data were recorded by CRV (formally Holland Genetics) from the OPU-IVP program from January 1995 to March 2006. Data were collected from 1508 Holstein female donors, both cows and pregnant virgin heifers, with a total of 18,702 OPU sessions. Data were analyzed with repeated-measure sire models with permanent environment effect using ASREML (Holstein Friesian). Estimates of heritability were 0.25 for Ncoc, 0.09 for Qcoc, 0.19 for NcleavD4, 0.21 for NembD7, 0.16 for NTembD7, 0.07 for PcleavD4, 0.12 for PembD7, and 0.10 for PTembD7. Genetic correlation between Ncoc and Qcoc was close to zero, whereas genetic correlations between Ncoc and the number of embryos were positive and moderate to high for NembD7 (0.47), NTembD7 (0.52), and NcleavD4 (0.85). Genetic correlations between Ncoc and percentages of embryos (PcleavD4, PembD7, and PTembD7) were all close to zero. Phenotypic correlations were in line with genetic correlations. Genetic and phenotypic correlations between Qcoc and all other traits were not significant except for the phenotypic correlations between Qcoc and number of embryos, which were negative and low to moderate for NembD7 (–0.20), NTembD7 (–0.24), and NcleavD4 (–0.43). Results suggest that cumulus-oocyte complex (COC) quality, based on cumulus investment, is independent from the total number of COCs collected via OPU and that in general, a higher number of COCs will lead to a higher number of embryos produced. The correlation between the estimated breeding values for Ncoc and PTembD7 of sires in this study and the sires breeding index for female-fertility based on the Dutch cattle population was close to zero. This study revealed OPU-IVP traits (NembD7, NTembD7, and Ncoc) that could be of potential value for selection. Introduction of such traits in breeding programs would enhance the number of offspring from superior donors as well as improve the cost efficiency of OPU-IVP programs.

The purposes of this paper were to 1) develop a stochastic model that would reflect observed variation between animals and across ages in immunocompetence and responsiveness; and 2) illustrate consequences of this variability for the statistical power of genotype comparisons and selection. A stochastic model of immunocompetence development and responsiveness kinetics was developed. This model enabled variability in immunological variables to be taken into account in the evaluation of challenge and measurement strategies for selection. The characteristics of the variation in model output reflect those observed in the literature, to the extent that variation in the literature shows a consistent pattern; knowledge of true variation and patterns of variation in immunological variables is limited. The model created correlations between immunocompetence and immunoresponsiveness components, as well as correlations within each component across time. These correlations were generally in agreement with literature estimates, where available. The model enabled predictions of the effectiveness of selection for improved health through immunocompetence or immunoresponsiveness. It was predicted that effective selection for increased general immunocompetence to improve health should be done only when baseline immunity has matured. Further, the model implied that selection is unlikely to be successful if it is based only on a single measurement. Problems with low statistical power to detect differences between genotypes can be reduced by increasing challenge age in the experimental design, and one should ensure that the effects of maternal immunity are minimal when the challenge is done. The ability to detect differences between different groups of animals differs substantially with measurement timing because of low repeatabilities of immunocompetence and responsiveness across time. In general, the probability of detecting differences becomes higher when the challenge age is increased. Consequently, both the age at selection and the age at which information is gathered for selection must be considered carefully when designing genetic evaluations.

The purpose of this study was two-fold: 1) to develop a deterministic model that describes the development of immunocompetence and the kinetics of immunoresponsiveness to a pathogenic challenge in chicks; and 2) to use this model to illustrate the importance of factors in experimental design, such as type of variable measured, measurement timing, and challenge age. Difficulties in evaluating immunological variables hinder attempts to improve animal health through selection on immunological variables. In young chicks, evaluating immunological variables is additionally complicated by immune system development and maternal immunity. The evaluation of immunocompetence and immunoresponsiveness, and the definition of appropriate challenge and measurement strategies, may be enabled through a mathematical model which captures the key components of the immune system and its development. Therefore, a model was developed that describes the development of immunocompetence as well as the kinetics of immunoresponsiveness to a pathogenic extra-cellular bacterial challenge in an individual chick from 0 to 56 days of age. The model consisted of four components describing immunocompetence (maternal- and baseline immunity) and immunoresponsiveness (acute phase- and antibody response). Individual component equations generally fitted published data adequatelly. Four scenarios that represented combinations of challenge age and measurement timing were simulated. In each scenario, the immunoresponsiveness to a particular challenge was compared for three different levels of baseline immunity, representing three broiler genotypes. It was illustrated that experimental design (type of immunoresponsiveness measured, measurement timing, and challenge age) can have an important effect on the ranking of genotypes, groups, or individuals, and on the reliability of extrapolations based on this ranking. It is concluded that this model is a potentially useful tool in the definition of appropriate challenge and measurement strategies when evaluating immunocompetence and immunoresponsiveness. Further, it may be used as a generator of hypotheses on global immunological relationships to be tested experimentally.

The purpose of this study was 2-fold: 1) to develop a deterministic model that describes the development of immunocompetence and the kinetics of immunoresponsiveness to a pathogenic challenge in chicks and 2) to use this model to illustrate the importance of factors in experimental design, such as type of variable measured, measurement timing, and challenge age. Difficulties in evaluating immunological variables hinder attempts to improve animal health through selection on immunological variables. In young chicks, evaluating immunological variables is additionally complicated by immune system development and maternal immunity. The evaluation of immunocompetence and immunoresponsiveness and the definition of appropriate challenge and measurement strategies may be enabled through a mathematical model that captures the key components of the immune system and its development. Therefore, a model was developed that describes the development of immunocompetence as well as the kinetics of immunoresponsiveness to a pathogenic extracellular bacterial challenge in an individual chick from 0 to 56 d of age. The model consisted of 4 components describing immunocompetence (maternal and baseline immunity) and immunoresponsiveness (acute phase and antibody response). Individual component equations generally fit published data adequately. Four scenarios that represented combinations of challenge age and measurement timing were simulated. In each scenario, the immunoresponsiveness to a particular challenge was compared for 3 different levels of baseline immunity, representing 3 broiler genotypes. It was illustrated that experimental design (type of immunoresponsiveness measured, measurement timing, and challenge age) can have an important effect on the ranking of genotypes, groups, or individuals and on the reliability of extrapolations based on this ranking. It is concluded that this model is a potentially useful tool in the definition of appropriate challenge and measurement strategies when evaluating immunocompetence and immunoresponsiveness. Further, it may be used as a generator of hypotheses on global immunological relationships to be tested experimentally.

The aims of this paper were two-fold: 1) to develop a stochastic model that reflects observed variation between animals and across age in immunocompetence and responsiveness; and 2) to illustrate consequences of this variability for statistical power of genotype comparisons and selection. The considerable amount of variability in immunological variables may cause non-normalities and heteroscedasticity in variance across age. A stochastic model of immunocompetence development and responsiveness kinetics was developed, to enable taking this variability into account in evaluation of challenge and measurement strategies for selection. The characteristics of the variation in model output reflect those observed in literature, to the extent that the variation in literature shows a consistent pattern. Actual knowledge of the true variation and patterns of variation in immunological variables is rather limited though. Due to causal relationships among the immunocompetence and responsiveness components, and due to the stochastic variation across time being correlated with an assumed lower limit of the repeatability, the model also creates correlations among these and within these across time. Correlations among components, as well as within components across time, were generally of similar size to estimates reported in literature. The model predictions on correlations among immune system components indicate that selection for improved health through improved immunocompetence or responsiveness may be difficult due to the sometimes low correlations among components as well as within components across time. The model also provides insight into experimental design. In practice, the statistical power of an experiment with a given challenge and measurement strategy to test a given hypothesis is often too low. The model predicts that problems with low power can be reduced by increasing challenge age in the experimental design, and that especially it should be confirmed or ensured that maternal immunity is absent when challenge is done. Repeatabilities of immunocompetence and responsiveness across time are also relatively low sometimes in practice. Consequently, the age at selection as well as the age at which information is gathered for selection must both be considered

The purpose of this study was to investigate whether differences in susceptibility to colibacillosis are associated with maternal antibodies, antibody response, and alterations in thyroid hormones [triiodothyronine (T3) and thyroxine (T4)] and to investigate the effect of genotype on the changes in T3 and T4 during challenge and antibody response. A challenge experiment was executed in 2 trials. Per trial, 24 chicks per genotype were challenged, and 20 chicks per genotype were controls. At 7 d of age, challenged chicks were intratracheally inoculated with 0.3 mL of Escherichia coli O78K80 and controls with 0.3 mL of PBS. All chicks were euthanized at 14 or 15 d. Thyroid hormone plasma concentrations and E. coli-specific antibody titers (AB) were measured at 7 d (T(3 d7), T(4 d7), and AB(d7)) and 14 or 15 d (change from 7 to 14 or 15 d was analyzed: DeltaT(3), DeltaT(4), and DeltaAB). Susceptibility was defined based on mortality, lesions, growth retardation, and eating behavior. There was a significant effect of challenge on T(3 d7); probably due to eating pattern in association with circadian rhythm. The challenge group was suggested to have functional hypothyroidism relative to the control group, indicating metabolic changes due to the challenge, and it was indicated that an antibody response was elicited. Differences in susceptibility were not significantly related to differences in T(3 d7), T(4 d7), DeltaT(3), or DeltaT(4) or to maternal antibodies (AB(d7)), but the antibody response tended to increase (decreasing DeltaAB) with increasing susceptibility. There were indications of genetic variation in T(4 d7), DeltaT(4), AB(d7), and DeltaAB, but there was no observed effect of genotype on DeltaT(3) and DeltaT(4) during challenge or on the antibody response. Further, there were indications that selection for growth traits has resulted in alterations in DeltaT(4) due to challenge, as indicated by a lower DeltaT(4) in the challenge group relative to the control group for more intensively selected genotypes as opposed to a higher DeltaT(4) for less intensively selected genotypes.