Dr. Kowalski's recent focus in methodological research is in defining
a new field of Inferential Statistical Bioinformatics that integrates
inference principles within a bioinformatics setting. Her emphasis
in the development of this field is on nonparametric (distribution-free)
approaches to facilitate the construction and tests of hypotheses
within the setting of very high-dimensional data relative to the
number of samples. Along these lines, Dr. Kowalski is developing
inference paradigms for analysis of high dimensional genetic and
genomic data through several avenues including but not limited to,
stochastic processes, modeling genetic heterogeneity, and composite
tests based on summary measures of heterogeneity (see below, Nonparametric
Methods for Genetic and Genomic Analysis).
Dr. Kowalski is currently under contract with Wiley to co-author
a book on the theory and applications of U-statistics for use as
part of a graduate school curriculum in Statistics and Biostatistics.
Her collaborative effort in this book is upon teaching the theory
and applications of U-statistics to address timely areas of genetic
and genomic statistical analyses.
Other research interests of Dr. Kowalski's include sequence analysis
for estimation of genetic diversity and recombination as it relates
to the Human Immunodeficiency Virus (HIV) genome. In this regard,
she has developed statistical methods for comparing and characterizing
genetic sequence heterogeneity associated with categorical phenotypes,
such as altered viral drug susceptibility to HIV. In addition to
genetic analyses, Dr. Kowalski has interests in measurement error
estimation for laboratory assay data and in extending generalized
estimating equations to accommodate such additional error.
