JABES Highlights
How Fast Does Rabies Spread? Examining
Crop Response to Irrigation
Carl Schwarz, JABES Editor, Simon Fraser University
H
ow fast does rabies spread interval of Newcombe is adapted. The methods are applied to a com-
over the landscape in a pop- parison of West Nile virus mosquito infection prevalence by trapping
ulation of wild animals? Is it height in field collections from Louisiana in 2003.
constrained by geographic features? Scott Holan and coauthors, in “Semiparametric Geographically
David Wheeler and Lance Waller, Weighted Response Curves with Application to Site-Specific
in “Mountains, Valleys, and Rivers: Agriculture,” develop a spatial model to examine crop response
The Transmission of Raccoon Rabies to irrigation under both spatial and treatment effects. A combina-
Over a Heterogeneous Landscape,” tion of models is used. A quadratic response curve forms the basis
develop a space-time diffusion model of relating crop-specific economic quantities to irrigation. A spa-
of a raccoon rabies outbreak across tially and treatment-varying coefficient model allows for spatial
several states in the eastern United variation or possible nitrogen-irrigation interactions. As is often
States. This work combines statisti- the case, these models were fit using a Bayesian spatial model. This
cal modeling and a geographic approach was used to model the mean response of corn to irrigation
information system to link observed amounts across a wide geographic area.
patterns of disease diffusion with Perfectly measured data is the exception, but most statisti-
local landscape values. Three analytical approaches are used: spatial cal models implicitly assume measurements are taken with small
prediction (kriging) is used to provide a descriptive pattern of the or no error. D. J. Taylor and coauthors, in “Statistical Models for
spread of the virus; Bayesian areal wombling is employed to detect Exposure-Biomarker Relationships with Measurement Error and
barriers for infectious disease transmission; and a hierarchical Censoring,” develop nonlinear mixed models and corresponding
Bayesian model with spatially varying coefficients is used to obtain likelihood expressions that can address several concerns in biomark-
model-based estimates of the impacts of spatial features. er data, such as errors in the measurement of true mean exposure
Mark-recapture researchers have developed a wide variety of and true mean biomarker levels, nonlinear exposure-biomarker rela-
experimental setups. Often, these methods have been developed tionships, background biomarker levels in unexposed individuals,
independently of each other. Matthew Schofield and Richard and response levels that fall below analytic limits of detection. They
Barker, in “A Unified Capture-Recapture Framework,” develop develop maximum likelihood estimation techniques that permit
a flexible hierarchical framework for capture-recapture data with valid statistical inferences to be made using standard software.
many of the current capture-recapture models included as spe- Michael Swartz and coauthors, in “Bayesian Variable Selection
cial cases. The hierarchical nature of the model also allows natural in Clustering High-Dimensional Data with Substructure,” exam-
expression of relationships, both between parameters and between ine the problem of variable selection in clustering micro-array data
parameters and the realization of random variables, such as pop- when there is additional structure (e.g., experimental design) a priori
ulation size. For example, existing methods cannot be used to present. Their method simultaneously determines which expression
model density dependence where survival and birth rates depend patterns are important and which genes contribute to such patterns
on the population size, but these models are easily handled in this in light of experimental groupings.
new framework. In plant and animal breeding, selection cycles consist of choosing
The binomial distribution has a long history in biometry and candidate individuals with high genotypic values for traits related
forms the basis of many statistical procedures, such as logistic regres- to observable phenotypic scores. Selection indexes used to select
sion. But what can be done if individual success/failure measures are the best individuals for the next breeding cycle are based on phe-
too expensive to collect, and pooling is used to reduce costs? Brad notypic observations of traits. The theory of a restrictive selection
Biggerstaff, in “Confidence Intervals for the Difference of Two index attempts to maximize the genetic progress of some characters
Proportions Estimated from Pooled Samples,” examines confidence while leaving others unchanged. Jesús Cerón-Rojas and coauthors,
intervals for the difference of two binomial proportions estimated in “A Restricted Selection Index Method Based on Eigenanalysis,”
from pooled samples with unequal pool sizes. Asymptotic methods generalize a particular type of selection index—the eigenanalysis
are used to derive Wald, profile score, and profile likelihood ratio method—to allow it to be used as a restrictive selection index. They
intervals. Corrections for bias and skewness are used to improve the applied it to a problem of doubled haploid maize mapping popula-
profile score interval. Further, the easily computed Wilson score-based tion of 236 genotypes with five traits. n
12 AMSTAT NEWS DECEMBER 2008
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82