This volume sets out a new statistical approach to test allelic association and genotype-specific effects in the genetic study of a disease. Primarily based on population-based association studies, the book also evaluates them compared to other methods.

The purpose of this book is to illustrate a new statistical approach to test allelic association and genotype-specific effects in the genetic study of diseases. There are some parametric and non-parametric methods available for this purpose. We deal with population-based association studies, but comparisons with other methods will also be drawn, analysing the advantages and disadvantages of each one, particularly with regard to power properties with small sample sizes. In this framework we will work out some nonparametric statistical permutation tests and likelihood-based tests to perform case-control analyses to study allelic association between marker, disease-gene and environmental factors. Permutation tests, in particular, will be extended to multivariate and more complex studies, where we deal with several genes and several alleles together. Furthermore, we show simulations under different assumptions on the genetic model and analyse real data sets by simply studying one locus with the permutation test.

Multivariate analysis in genetic studies Powerful and flexible statistical solutions Fully nonparametric analysis

Autorentext
Prof. Luigi Salmaso: Study period (1995) with an ERASMUS grant at the Department of Mathematics - Statistics Division of Athens University. Degree in Statistical and Economic Science from University of Padova. Graduation thesis developed under the supervision of Prof. Stratis Kounias (Dep. of Mathematics - Statistics Division of Athens University). Study period at Department of Mathematics - Statistics Division of Athens University (1998). Study period (1999) at the following universities and research centers of the United States: Texas Tech University (Department of Mathematics and Statistics), Kansas State University (Department of Statistics), University of Washington (Department of Statistics), MathSoft (Seattle branch). Ph.D. in Statistics, from University of Padova with Thesis entitled: "Orthogonal two-level factorial designs and permutation testing for effects", developed under the supervision of Prof. Fortunato Pesarin (Dep. of Statistical Sciences - University of Padova) and Prof. Stratis Kounias (Dep. of Mathematics, Athens University). Since 1996 collaboration with the Consorzio Mario Negri Sud - Pharmacological and Biomedical Research Center, concerning the development and the application of nonparametric multivariate methods for hypothesis testing in biomedical studies. Since 1999 member of MURST project: Methods of Statistic Inference for Complex Problems, coordinated by Prof. F. Pesarin (Dep. of Statistical Sciences, University of Padova). Teacher of Statistics at the Degree Course in Managerial Engineering of the Engineering Faculty of University of Padova. Teacher of short refresher courses about nonparametric methods at the Pharmacological Research Centre Consorzio Mario Negri Sud (Chieti).

Inhalt
Introduction.- Association Studies in Genetics.- The Nonparametric Permutation Methodology.- Statistical Problems of Allelic Association.- Power and Sample Size Simulations.- Case Study.- Conclusions.- References.