Four statistical models for genetic evaluations utilising dairy test day data are considered. These are: the fixed regression model, the random regression model, the autoregressive model and a multiple trait model. The emphasis is put on the comparison of these models in terms of their assumed covariance structure, modelling and prediction of breeding values and parameterisation. In the future one of the models should be used for a routine genetic evaluation of the Polish Black-and-White dairy cattle. Therefore, characteristics of test day data from the Polish population are given. In conclusion, it appears that thanks to its flexibility in handling heterogeneous variances during lactation, variable autocorrelation and non-uniform spacing between tests, the random regression model forms the most suitable approach.
The primary goal of this study was to investigate statistical properties of a mixed inheritance model for the localization of quantitative trait loci (QTL). This is based on the analysis of phenotypic data for the amount of intramuscular fat (IMF) scored on 305 individuals originating from a cross between Duroc and Norwegian Landrace breeds. Marker genotype information is available for F1 and F2 generations. Statistical procedures compared involve i) the interval mapping, ii) the composite interval mapping, iii) a regression method, and iv) a mixed inheritance model accounting for a random animal additive genetic effect and relationships between individuals. The basic statistical properties of the latter approach are then assessed using Monte Carlo simulations showing slight unconservativeness as compared to and reasonable power to detect QTL of moderate effects. In the analysis of IMF data, the significant evidence for the existing QTL is detected on chromosome 6. A chromosomal region recommended for a second-step fine mapping analysis is identified between markers SW1823 and S0228, based on three types of confidence intervals derived by using: i) the Jackknife algorithm, ii) the numerical variance approximation, and iii) the LOD score approach. The Jackknife algorithm was additionally used to quantify each family?s contribution to the test statistic and to the estimate of QTL position.
The transmission-disequilibrium test (TDT) is a model-free method to detect linkage between a marker and a trait locus. Originally developed to map disease genes in human genetics, this statistic has been recently extended to deal with quantitative characters. The emphasis of current research is on investigating statistical properties of the test applied to data from livestock populations. For various constellations of sample parameters, it is shown via simulation that the empirically derived null hypothesis distribution of TDT remains in good agreement with its asymptotic distribution while its power is satisfactory only for very close linkage. TDT is then applied to a real data set from milk production data of a dairy cattle population.
Gas cleaning using plasma technology is slowly introduced into industry nowadays. Several challenges still have to be overcome: increasing the scale, safety, life-time and reducing costs. In 2006 we demonstrated a 20 kW nanosecond pulsed corona system. The electrical efficiency was > 90%. O-radical yields were found to be very high (3-7 mole/kWh). However, to be competitive, high costs of the pulsed power technology are still a major hurdle. Here we present a novel modulator for efficient generation of large volume corona plasma. Only a small amount of expensive high-voltage components are required. Switching is done at an intermediate voltage level of 1 kV with standard thyristors. At the high-voltage side, only a diode and a pulse transformer are needed. The estimated costs are about 5 kEuro/kW, whereas for state-of-the-art pulsed power technology these costs usually are about 20-30 kEuro/kW. Detailed investigations on the modulator and a wire-plate corona reactor will be presented. Modulator parameters have been varied systematically as well as reactor parameters (number of electrodes, electrode-plate distance). The O-radical yield was determined from the measured ozone concentrations at the exhaust of the reactor. With a detailed kinetic model, ozone concentrations could be calculated back to the initial O*-yields. The following conclusions will be discussed: for all parameters, an electrical efficiency of > 90% could be obtained. With fast imaging, the average streamer width was found to be ∼ 737 μm and an estimate for the plasma volume was made. The obtained yields of O-radicals (1-4 mole/kWh) are excellent. The conditions to obtain high yields will be discussed.
For one kind of high current diodes composed of a ceramic-metal welding vacuum interface, the electrical design was presented. For compactness, a radial type insulator and a cone-column anode crust were adopted. The shielding methods around cathode and anode region were applied to mitigate the influence of welding solder to vacuum flashover. Finite Element Analysis (FEA) simulation results indicated that by adjusting the anode outline and shielding shape, the electric fields along the ceramic were well distributed. High voltage test was conducted on a long-pulse accelerator and experimental results confirm the theoretic design: the diode can stably hold on 400 kV and 200 ns voltage pulse.
We analysed data from a selective DNA pooling experiment with 130 individuals of the arctic fox (Alopex lagopus), which originated from 2 different types regarding body size. The association between alleles of 6 selected unlinked molecular markers and body size was tested by using univariate and multinomial logistic regression models, applying odds ratio and test statistics from the power divergence family. Due to the small sample size and the resulting sparseness of the data table, in hypothesis testing we could not rely on the asymptotic distributions of the tests. Instead, we tried to account for data sparseness by (i) modifying confidence intervals of odds ratio; (ii) using a normal approximation of the asymptotic distribution of the power divergence tests with different approaches for calculating moments of the statistics; and (iii) assessing P values empirically, based on bootstrap samples. As a result, a significant association was observed for 3 markers. Furthermore, we used simulations to assess the validity of the normal approximation of the asymptotic distribution of the test statistics under the conditions of small and sparse samples.