A Hybrid Method of Multidimensional Scaling and Clustering for Determining Genetic Influence on Phenotypes

Authors:
Qiao Li;Wenjia Wang;Alexander J. Macgregor;George Smith
Affiliations:
School of Computing Sciences, University of East Anglia, Norwich, UK NR4 7TJ;School of Computing Sciences, University of East Anglia, Norwich, UK NR4 7TJ;School of Medicine, University of East Anglia, Norwich, UK NR4 7TJ;School of Computing Sciences, University of East Anglia, Norwich, UK NR4 7TJ
Venue:
ADMA '09 Proceedings of the 5th International Conference on Advanced Data Mining and Applications
Year:
2009

Citing 2
Cited 0

Data clustering: a review

ACM Computing Surveys (CSUR)
Data Mining: Concepts, Models, Methods and Algorithms

Data Mining: Concepts, Models, Methods and Algorithms

Quantified Score

Hi-index	0.00

Visualization

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.