Related Experiment Videos
Data mining and computationally intensive methods: summary of Group 7 contributions to Genetic Analysis Workshop 13
Tracy J Costello1, Catherine T Falk, Kenny Q Ye
1Department of Epidemiology, University of Texas M.D. Anderson Cancer Center, Houston, USA.
Genetic Epidemiology
|November 25, 2003
Summary
Data-mining methods were applied to Framingham Heart Study data to uncover genetic factors for cardiovascular disease. These flexible techniques identified complex genetic and environmental influences on health traits.
Area of Science:
- Genetics
- Biostatistics
- Computational Biology
Background:
- The Framingham Heart Study dataset provides a well-characterized resource for investigating complex diseases.
- Understanding the genetic and environmental contributions to cardiovascular disease is crucial for public health.
Purpose of the Study:
- To test and compare novel statistical methodologies for genetic analysis.
- To elucidate the contributions of genes, environment, and their interactions to cardiovascular disease and related traits.
Main Methods:
- Application of data-mining methodologies including tree-based methods, neural networks, discriminant analysis, and Bayesian variable selection.
- Utilizing both Framingham Heart Study data and a related simulated dataset for method validation.
Main Results:
- Data-mining strategies demonstrated flexibility and potential in identifying factors involved in complex disorders.
- Investigations aimed to identify underlying genetic factors for cardiovascular disease and related traits.
Conclusions:
- Data-mining approaches offer powerful tools for dissecting complex genetic and environmental interactions in disease.
- The study highlights the utility of diverse statistical methods in genetic analysis and disease research.