POISSON COKRIGING AS A GENERALIZED LINEAR MIXED MODEL.

It is often of interest to predict spatially correlated count outcomes that follow a Poisson distribution. For example, in the environmental sciences we may want to predict pollen counts using temperature or precipitation data as auxiliary variables. To predict a Poisson outcome variable in the presence of an auxiliary variable, Poisson cokriging as a Generalized Linear Mixed Model (GLMM) is proposed. This model has a bivariate structure with a Poisson outcome variable and an auxiliary variable. A covariance matrix similar to that used in cokriging is assumed. A simulation study and a real data example using the number of microplastics in the digestive tracts of fish are presented. The results showed that Poisson cokriging methodology can be applied successfully in practice with small average errors and coverage close to 95%. The Poisson cokriging model can be a useful tool for spatial prediction.

Determining the Efficacy of a Hybridizing Agent in Wheat (Triticum aestivum L.).

Hybrid wheat (Triticum spp.) has the potential to boost yields and enhance production under changing climates to feed the growing global population. Production of hybrid wheat seed relies on male sterility, the blocking of pollen production, to prevent self-pollination. One method of preventing self-pollination in the female plants is to apply a chemical hybridizing agent (CHA). However, some combinations of CHA and genotypes have lower levels of sterility, resulting in decreased hybrid purity. Differences in CHA efficacy are a challenge in producing hybrid wheat lines for commercial and experimental use. Our primary research questions were to estimate the levels of sterility for wheat genotypes treated with a CHA and determine the best way to analyze differences. We applied the CHA sintofen (1-(4-chlorphyl)-1,4-dihydro-5-(2-methoxyethoxy)-4-oxocinnoline-3-carboxylic acid; Croisor 100) to 27 genotypes in replicate. After spraying, we counted seed in bagged female heads to evaluate CHA efficacy and CHA-by-genotype interaction. Using logit and probit models with a threshold of 7 seeds, we found differences among genotypes in 2015. Sterility was higher in 2016 and fewer genotypic differences were found. When CHA-induced sterilization is less uniform as in 2015, zero-inflated and hurdle count models were superior to standard mixed models. These models calculate mean seed number and fit data with limit-bounded scales collected by agronomists and plant breeders to compare genotypic differences. These analyses can assist in selecting parents and identifying where additional optimization of CHA application needs to occur. There is little work in the literature examining the relationship between CHAs and genotypes, making this work fundamental to the future of hybrid wheat breeding.

Spatial variation in western corn rootworm (Coleoptera: Chrysomelidae) susceptibility to Cry3 toxins in Nebraska.

Repeated use of field corn (Zea mays L.) hybrids expressing the Cry3Bb1 and mCry3A traits in Nebraska has selected for field-evolved resistance in some western corn rootworm (WCR; Diabrotica virgifera virgifera LeConte) populations. Therefore, this study was conducted to characterize spatial variation in local WCR susceptibility to Cry3Bb1 and mCry3A traits in Keith and Buffalo counties, Nebraska, and determine the relationship between past management practices and current WCR susceptibility. Adult WCR populations were collected from sampling grids during 2015 and 2016 and single-plant larval bioassays conducted with F1 progeny documented significant variation in WCR susceptibility to Cry3Bb1 and mCry3A on different spatial scales in both sampling grids. At the local level, results revealed that neighboring cornfields may support WCR populations with very different susceptibility levels, indicating that gene flow of resistant alleles from high trait survival sites is not inundating large areas. A field history index, comprised of additive and weighted variables including past WCR management tactics and agronomic practices, was developed to quantify relative selection pressure in individual fields. The field history index-Cry3 trait survivorship relationship from year 1 data was highly predictive of year 2 Cry3 trait survivorship when year 2 field history indices were inserted into the year 1 base model. Sensitivity analyses indicated years of trait use and associated selection pressure at the local level were the key drivers of WCR susceptibility to Cry3 traits in this system. Retrospective case histories from this study will inform development of optimal resistance management programs and increase understanding of plant-insect interactions that may occur when transgenic corn is deployed in the landscape. Results from this study also support current recommendations to slow or mitigate the evolution of resistance by using a multi-tactic approach to manage WCR densities in individual fields within an integrated pest management framework.

Short communication: On recognizing the proper experimental unit in animal studies in the dairy sciences.

Sound design of experiments combined with proper implementation of appropriate statistical methods for data analysis are critical for producing meaningful scientific results that are both replicable and reproducible. This communication addresses specific aspects of design and analysis of experiments relevant to the dairy sciences and, in so doing, responds to recent concerns raised in a letter to the editor of the Journal of Dairy Science regarding journal policy for research publications on pen-based animal studies. We further elaborate on points raised, rectify interpretation of important concepts, and show how aspects of statistical inference and elicitation of research conclusions are affected.

Pattern Recognition of Longitudinal Trial Data with Nonignorable Missingness: An Empirical Case Study.

Methods for identifying meaningful growth patterns of longitudinal trial data with both nonignorable intermittent and drop-out missingness are rare. In this study, a combined approach with statistical and data mining techniques is utilized to address the nonignorable missing data issue in growth pattern recognition. First, a parallel mixture model is proposed to model the nonignorable missing information from a real-world patient-oriented study and concurrently to estimate the growth trajectories of participants. Then, based on individual growth parameter estimates and their auxiliary feature attributes, a fuzzy clustering method is incorporated to identify the growth patterns. This case study demonstrates that the combined multi-step approach can achieve both statistical gener ality and computational efficiency for growth pattern recognition in longitudinal studies with nonignorable missing data.