No abstract

To accelerate the molecular analysis of behavior in the honey bee (Apis mellifera), we created expressed sequence tag (EST) and cDNA microarray resources for the bee brain. Over 20,000 cDNA clones were partially sequenced from a normalized (and subsequently subtracted) library generated from adult A. mellifera brains. These sequences were processed to identify 15,311 high-quality ESTs representing 8912 putative transcripts. Putative transcripts were functionally annotated (using the Gene Ontology classification system) based on matching gene sequences in Drosophila melanogaster. The brain ESTs represent a broad range of molecular functions and biological processes, with neurobiological classifications particularly well represented. Roughly half of Drosophila genes currently implicated in synaptic transmission and/or behavior are represented in the Apis EST set. Of Apis sequences with open reading frames of at least 450 bp, 24% are highly diverged with no matches to known protein sequences. Additionally, over 100 Apis transcript sequences conserved with other organisms appear to have been lost from the Drosophila genome. DNA microarrays were fabricated with over 7000 EST cDNA clones putatively representing different transcripts. Using probe derived from single bee brain mRNA, microarrays detected gene expression for 90% of Apis cDNAs two standard deviations greater than exogenous control cDNAs.

In this paper, we propose a flexible "two-part" random Effects model (Olsen and Schafer 2001;Tooze, Grunwald, and Jones 2002) for correlated medical cost data. Typically, medical cost data are right-skewed, involve a substantial proportion of zero values, and may exhibit heteroscedasticity. In many cases, such data is also obtained in hierarchical form, e.g., on patients served by the same physician. The proposed model specification therefore consists of two generalized linear mixed models (GLMM), linked together by correlated random Effects. Respectively, and conditionally on the random Effects and covariates, we model the odds of cost being positive (Part I) using a GLMM with a logistic link and the mean cost (Part II) given that costs were actually incurred using a generalized gamma regression model with random Effects and a scale parameter that is allowed to depend on covariates (c.f. Manning, Basu, and Mullahy 2005). The class of generalized gamma distributions is very flexible and includes the lognormal, gamma, inverse gamma and Weibull distributions as special cases. We demonstrate how to carry out estimation using the Gaussian quadrature techniques conveniently implemented in SAS Proc NLMIXED. The proposed model is used to analyze pharmacy cost data on 56,245 adult patients clustered within 239 physicians in a mid-western U.S. managed care organization.

PSNHDD appears to be a clinically informative end point measure, especially when used with a grace period, and is as sensitive as most traditional outcome measures in detecting differences between the medication and placebo groups. Nonetheless, these findings should be replicated in other clinical data sets, particularly with medications that work via different mechanisms.