Statistical Analysis of Linkage/Association on Family-Based Studies in Human Genetics

Columbia University, New York NY

Investigators

Abstract

The allelic association between disease genes and marker alleles allows predicting which marker allele is in coupling with the disease. There are novel statistical testing procedures proposed recently. For example, the Transmission/ Disequilibrium Testing ( TDT ) has been both a popular and powerful method to evaluate the linkage/ association between the candidate genes with disease and the markers. This approach combines the information of association and linkage to yield greater power than conventional tests for detecting linkage when the association is present. It also has great potential to play a major role in future gene mapping of human disorder, by carrying out genomewide TDT screens. Recent literature has seen an increasing trend in the number of studies that have applied TDT or its generalizations to a variety of complex family-based data, but these methods are ad hoc. There is a great need for a systematic and sensible optimal analysis of these methods. This work sets up a general statistical framework that can incorporate all procedures, different data sets and other related knowledge such as disease information, sampling designs and population information. We develop a general statistical theory, useful as a guideline for future investigators, who can design and develop their own statistical procedures (following the framework) to meet their needs in incorporating different environments. The research provides a likelihood-based approach by introducing a useful 3x2 table which applies to each nuclear family. It is natural to adopt the conditionality principle to derive and to evaluate a variety of TDT type procedures and linkage tests, using "Locally Most Powerful Testing" (LMPT ) as a central optimality criteria. This research outlines the likelihood approach and demonstrates how the method works in some special cases. The research provides the derivations and findings from the examples,which should shed light on a better understanding of the more challenging and broader problems arising from these data sets. In training doctoral students working on this increasingly important area, the research provides the basis to develop a new research level course. The proposed research intends to lead ultimately to routine guidelines for future investigators and to offer a theoretical basis for family-based studies.