A hybrid method of multidimensional scaling and clustering for determining genetic influence on phenotypes

Abstract

As a branch of modern biomedical study, genetic epidemiology research on complex diseases usually aims to identify the genetic influences on phenotypes. This paper presents a hybrid method named MDS-C, which combines the multidimensional scaling method and a clustering algorithm, to unveil the genetic relationships among phenotypes by using phenotypic information only. In MDS-C, the cross-twin cross-trait correlation between any two phenotypes is designed to measure the genetic similarity. MDS-C is verified by a series of simulation studies. Then it is applied to a real bone mineral density (BMD) dataset collected by the St Thomas' UK Adult Twin Registry. Its results suggest that the genetic influence on BMD is site-specific. © 2009 Springer.