Maternal, dominance and additive genetic effects in Nile tilapia; influence on growth, fillet yield and body size traits.

There are only few studies of dominance effects in non-inbred aquaculture species, since commonly used mating designs often have low power to separate dominance, maternal and common environmental effects. Here, a factorial design with reciprocal cross, common rearing of eggs and subsequent lifecycle stages and pedigree assignment using DNA microsatellites was used to separate these effects and estimate dominance (d2) and maternal (m2) ratios in Nile tilapia for six commercial traits. The study included observations on 2524 offspring from 155 full-sib families. Substantial contributions of dominance were observed (P < 0.05) for body depth (BD) and body weight at harvest (BWH) with estimates of d2 = 0.27 (s.e. 0.09) and 0.23 (s.e. 0.09), respectively in the current breeding population. In addition the study found maternal variance (P < 0.05) for BD, BWH, body thickness and fillet weight explaining ~10% of the observed phenotypic variance. For fillet yield (FY) and body length (BL), no evidence was found for either maternal or dominance variance. For traits exhibiting maternal variance, including this effect in evaluations caused substantial re-ranking of selection candidates, but the impact of including dominance effects was notably less. Breeding schemes may benefit from utilising maternal variance in increasing accuracy of evaluations, reducing bias, and developing new lines, but the utilisation of the dominance variance may require further refinement of parameter estimates.

Conspicuous carotenoid-based pelvic spine ornament in three-spined stickleback populations-occurrence and inheritance.

Reports on reddish carotenoid-based ornaments in female three-spined sticklebacks (Gasterosteus aculeatus) are few, despite the large interest in the species' behaviour, ornamentation, morphology and evolution. We sampled sticklebacks from 17 sites in north-western Europe in this first extensive study on the occurrence of carotenoid-based female pelvic spines and throat ornaments. The field results showed that females, and males, with reddish spines were found in all 17 populations. Specimens of both sexes with conspicuous red spines were found in several of the sites. The pelvic spines of males were more intensely red compared to the females' spines, and large specimens were more red than small ones. Fish infected with the tapeworm (Schistocephalus solidus) had drabber spines than uninfected fish. Both sexes had red spines both during and after the spawning period, but the intensity of the red colour was more exaggerated during the spawning period. As opposed to pelvic spines, no sign of red colour at the throat was observed in any female from any of the 17 populations. A rearing experiment was carried out to estimate a potential genetic component of the pelvic spine ornament by artificial crossing and rearing of 15 family groups during a 12 months period. The results indicated that the genetic component of the red colour at the spines was low or close to zero. Although reddish pelvic spines seem common in populations of stickleback, the potential adaptive function of the reddish pelvic spines remains largely unexplained.

Genetic analysis of vibriosis and viral nervous necrosis resistance in Atlantic cod (Gadus morhua L.) using a cure model.

The aim of this study was to investigate whether observed time-until-death of Atlantic cod (Gadus morhua L.) juveniles in separate challenge tests with Vibrio anguillarum (causes vibriosis) and nodavirus [causes viral nervous necrosis (VNN)] are due to differences in susceptibility (whether at risk or not) or increased endurance (individual hazard, given that the animal is susceptible) using a cure mixture (CURE) model with Gibbs sampling. Observed time-until-death, prepared as sequential binary records, were analyzed with the CURE model and results were compared with cross-sectional threshold (SIMPLE) and an ordinary longitudinal survival score (NAÏVE) model (i.e., assuming that all animals are susceptible). Overall mortality at the end of the test was 86 and 71% for vibriosis and VNN, respectively. But the CURE model estimated 92 and 82% of the population to be susceptible to vibriosis and VNN, respectively. Hence, a substantial fraction among the survivors were considered to be susceptible but with high endurance. The underlying heritability of susceptibility was moderate for vibriosis (0.33) and extremely high for VNN (0.91), somewhat greater compared with classical SIMPLE model (0.19 and 0.76 for vibriosis and VNN, respectively), analyzing end survival as a cross-sectional binary trait. Estimates of the underlying heritability were low for single test-day scores of both endurance (0.02 and 0.15 for vibriosis and VNN, respectively) in the CURE model and for the NAÏVE model (0.02 and 0.18 for vibriosis and VNN, respectively). Based on the CURE model, the genetic correlation between susceptibility and endurance was low to moderately positive and significantly different from unity (P < 0.01) for both vibriosis (0.13) and VNN (0.47). Estimated breeding values from the SIMPLE and NAÏVE models showed moderate to high correlations (0.41 to 0.96) with EBV for susceptibility and endurance in the CURE model. The analyses indicate that susceptibility and endurance are apparently distinct genetic traits. Still, the genetic variation estimated in the SIMPLE and NAÏVE models seems to a large extent to be controlled by susceptibility and an efficient genetic selection for reduced susceptibility to vibriosis and VNN is therefore likely feasible even when using classical (noncure) models. Earlier termination of the challenge test or back truncation of survival data is not recommended as this likely shifts the focus of selection towards endurance rather than susceptibility.

Consistent boldness behaviour in early emerging fry of domesticated Atlantic salmon (Salmo salar): Decoupling of behavioural and physiological traits of the proactive stress coping style.

Individual variation in the way animals cope with stressors has been documented in a number of animal groups. In general, two distinct sets of behavioural and physiological responses to stress have been described: the proactive and the reactive coping styles. Some characteristics of stress coping style seem to be coupled to the time to emerge of fry from spawning redds in natural populations of salmonid fishes. In the present study, behavioural and physiological traits of stress coping styles were compared two and five months after emergence in farmed Atlantic salmon (Salmo salar), using individuals with an early or late time to emerge. Initially, compared to late emerging individuals, early emerging individuals showed a shorter time to resume feeding after transfer to rearing in isolation. Resumption of feeding after isolation was suggested to be related to boldness behaviour, rather than hunger, in the present study. This observation was repeated five months after emergence, demonstrating behavioural consistency over time in this trait. However, in other traits of proactive and reactive stress coping styles, such as social status, resting metabolism or post stress cortisol concentrations, early and late emerging individuals did not differ. Therefore, this study demonstrates that boldness in a novel environment is uncoupled from other traits of the proactive and reactive stress coping styles in farmed salmonids. It is possible that this decoupling is caused by the low competitive environment in which fish were reared. In natural populations of salmonids, however, the higher selection pressure at emergence could select for early emerging individuals with a proactive coping style.

Efficiency of estimation of variance and covariance components from full-sib group means for continuous or binary records.

SUMMARY: High fecundity in fish allows for testing of large full-sib groups and testing of sub-samples of each group for different traits in different environments, and this benefit is especially useful when traits cannot be measured on the breeding candidate itself. A MIVQUE-method that compresses the information in such data, by estimating variance components based on full-sib mean data, is presented for a two-trait model. The efficiency of this method is compared in a simulation study with a model using an individual observation vector when estimating additive genetic and residual (co)variances. The effect on efficiency of continuous versus binary traits, of varied average number of full-sibs, of varied frequencies in the binary traits, of different combinations of frequencies in the two binary traits, of different heritabilities, of different genetic correlations, and of two different design sizes were studied. Generally, all parameters influenced the efficiency, and this influence was larger on the residual, than on the additive genetic variance component. The proposed methodology will be especially well-suited for analysing binary traits because the residual variance component can be calculated via the phenotypic variance, which can be estimated on basis of the frequency of the trait. ZUSAMMENFASSUNG: Effizienz der Schaetzung von Varianz- und Kovarianzkomponenten aus Vollgeschwistergruppen-Mitelwerten bei kontinuierlichen und dichotomen Merkmalen Die hohe Vermehrungsrate von Fischen erlaubt es, große Vollgeschwistergruppen zu pruefen und auch Stichproben aus diesen für verschiedene Merkmale in verschiedenen Umwelten zu testen. Dies ist besonders hilfreich, wenn die interessierenden Merkmale nicht an den zu selektierenden Individuen erhoben werden können. Es wird ein MIVQUE-Ansatz zur Varianz- und Kovarianzkomponentenschätzung fuer ein Zwei-Merkmalsmodell vorgestellt, der auf der Beruecksichtigung von Vollgeschwistergruppen-Mittelwerten beruht. Die Effizienz dieses Ansatzes wird in einer Simulationsstudie untersucht, indem dieser mit einem auf individuellen Leistungen basierenden Ansatz verglichen wird. Dabei wird untersucht, welche Unterschiede sich bei kontinuierlichen und dichotomen Variablen ergeben sowie der Einfluß verschiedener Gruppengroessen, verschiedener Frequenzen der dichotomen Variablen, unterschiedlicher Heritabilitaeten und genetischer Korrelationen sowie zweier unterschiedlicher Designs. Im Allgemeinen waren die Einfluesse auf die geschaetzte Restvarianz immer groesser als die auf die genetischen Varianzkomponenten. Die vorgeschlagene Methode erwies sich als besonders geeignet fuer die Auswertung dichotomer Merkmale, weil dort die Groesse der Restvarianz sehr einfach aus der phaenotypischen Varianz approximiert werden kann, welche wiederum aus der beobachteten Frequenz des Merkmals gut zu schaetzen ist.