@article{wang_genton_2006, title={The multivariate skew-slash distribution}, volume={136}, ISSN={["1873-1171"]}, DOI={10.1016/j.jspi.2004.06.023}, abstractNote={The slash distribution is often used as a challenging distribution for a statistical procedure. In this article, we define a skewed version of the slash distribution in the multivariate setting and derive several of its properties. The multivariate skew-slash distribution is shown to be easy to simulate from and can therefore be used in simulation studies. We provide various examples for illustration.}, number={1}, journal={JOURNAL OF STATISTICAL PLANNING AND INFERENCE}, author={Wang, J and Genton, MG}, year={2006}, month={Jan}, pages={209–220} }
 @article{wang_boyer_genton_2004, title={A note on an equivalence between chi-square and generalized skew-normal distributions}, volume={66}, ISSN={["0167-7152"]}, DOI={10.1016/j.spl.2003.11.006}, abstractNote={In this note, we establish an equivalence between chi-square and generalized skew-normal distributions. This result is based on a distributional invariance property of even functions in generalized skew-normal random vectors. It extends the chi-square properties related to univariate and multivariate skew-normal distributions.}, number={4}, journal={STATISTICS & PROBABILITY LETTERS}, author={Wang, JZ and Boyer, J and Genton, MG}, year={2004}, month={Mar}, pages={395–398} }
 @article{groot_ward_wang_pokrzywa_o’brien_bennett_kelly_santangelo_schal_gould_2004, title={Introgressing pheromone QTL between species: Towards an evolutionary understanding of differentiation in sexual communication}, volume={30}, ISSN={["1573-1561"]}, DOI={10.1007/s10886-004-7946-y}, abstractNote={As a first step toward understanding how noctuid moths evolve species-specific pheromone communication systems, we hybridized and back-crossed two closely related moth species, Heliothis virescens (Hv) and H. subflexa (Hs), which differ qualitatively and quantitatively in their multicomponent sex pheromone blends. We used amplified fragment length polymorphism (AFLP) marker-based mapping of backcross families to determine which of the 30 autosomes in these moths contained quantitative trait loci (QTL) controlling the percentages of specific chemical components in the pheromone blends. In two previous backcrosses to Hs, we found a strong depressive effect of Hv-chromosome 22 on the percentage of three acetate components in the pheromone gland. These acetates are present in Hs and absent in Hv. Here, we describe how we introgressed Hv-chromosome 22 into the genomic background of Hs. Selection for Hv-chromosome 22 started from backcross 3 (BC3) females. All females that had Hv-chromosome 22 and a low percentage of acetates (<3% of the total amount of pheromone components present) were backcrossed to Hs males. In BC5 to BC8, we determined whether Hv-chromosome 22 was present by a) running only the primer pairs that would yield the markers for that chromosome, and/or b) determining the relative percentages of acetates in the pheromone glands. Either or both genotype and phenotype were used as a criterion to continue to backcross these females to Hs males. In BC9, we confirmed the isolation of Hv-chromosome 22 in the Hs genomic background, and backcrossed the males to Hs females to eliminate the Hv-sex chromosome as well as mitochondrial DNA. The pheromone composition was determined in BC3, BC5, and BC11 females with and without Hv-chromosome 22. All backcross females with Hv-chromosome 22 contained significantly less acetates than females without this chromosome. In addition, BC3 females with Hv-chromosome 22 contained significantly more Z11-16:OH than BC3 females without Hv-chromosome 22. However, in BC5 and BC11 females, the correlation between Z11-16:OH and Hv-chromosome 22 was lost, suggesting that there are separate QTL for the acetates and for Z11-16:OH, and that the relative amount of the alcohol component is only affected in epistasis with other (minor) QTL. Now that we have succeeded in isolating the chromosome that has a major effect on acetate production, we can test in behavioral experiments whether the presence of acetates may have been a driving force for a shift in pheromone composition. Such tests are necessary to move towards an evolutionary understanding of the differentiation in sexual communication in Heliothis spp. moths.}, number={12}, journal={Journal of Chemical Ecology}, author={Groot, A.T. and Ward, C. and Wang, J. and Pokrzywa, A. and O’Brien, J. and Bennett, J. and Kelly, J. and Santangelo, R.G. and Schal, C. and Gould, F.}, year={2004}, pages={2495–2514} }
 @article{barbercheck_wang_brownie_2003, title={Adaptation of the entomopathogenic nematode, Steinernema carpocapsae, to insect food plant}, volume={27}, ISSN={["1090-2112"]}, DOI={10.1016/S1049-9644(02)00187-1}, abstractNote={Abstract We conducted assays to determine if characteristics of entomopathogenic nematodes vary in response to continued exposure to a particular insect/host plant combination, and whether selection results in changed performance on other insect/host plant combinations. Three isolates of Steinernema carpocapsae (Weiser) (Agriotos, Mexican, and Hybrid) were continuously cultured in corn-fed ( Zea mays L.) or squash-fed ( Cucurbita pepo L.) southern corn rootworm, Diabrotica undecimpunctata howardi Barber for 25 passages. The rootworm-selected nematodes were compared to the same isolates maintained on Galleria mellonella (L.) (“unselected”). There was no increase in the ability to kill rootworms in selected nematodes compared to unselected nematodes. Among rootworm-selected nematodes, there were changes based on the host plant on which the rootworms had fed. The ability of nematodes selected on corn-fed rootworms to kill corn-fed rootworms compared to squash-fed rootworms improved in the Mexican and Agriotos isolates. The squash-selected Mexican isolate lost virulence against rootworms. Virulence of the Hybrid isolate did not change in response to selection on rootworms or host plant/rootworm combination. In general, changes in infective juvenile (IJ) production were not related to number of passages through rootworms. Using computer-aided image analysis, we measured the effects of selection on length and neutral lipid density and area in IJ. IJ selected on corn-fed rootworms were longer than those selected on squash-fed rootworms or maintained in G. mellonella . Length of IJ was positively correlated with the area of stained neutral lipids. The Hybrid isolate had a greater area and density of stained lipids than the Agriotos or Mexican isolates. IJ from nematodes selected on corn-fed rootworms tended to have a greater area of stained lipids, but not lipid density, compared with IJ selected on squash-fed rootworms or maintained in G. mellonella . There was a pattern for an inverse relationship between numbers of IJ produced and measures of lipid content. This pattern is discussed with reference to trade-offs in life history characteristics. Our results suggest that some characteristics of entomopathogenic nematodes can respond to selection on insect host plant, but that changes are dependent on the nematode isolate and the particular host plant on which the insect feeds.}, number={1}, journal={BIOLOGICAL CONTROL}, author={Barbercheck, ME and Wang, J and Brownie, C}, year={2003}, month={May}, pages={81–94} }