Presentation Title

Bayesian Clustering and Variable Selection for Gene Expression Data

Faculty Mentor

Valerie Poynor

Start Date

23-11-2019 8:00 AM

End Date

23-11-2019 8:45 AM

Location

247

Session

poster 1

Type of Presentation

Poster

Subject Area

physical_mathematical_sciences

Abstract

Gene Expression data describes the activity level of genes in a particular cell type. All cells in an organism will consist of the same genes, however, depending on the cell type, the gene activity may vary. We consider a dataset describing the gene expressions of five different cancer types. We are interested in clustering the cancer types according to the various gene activity. Statistically, this is a challenge due to the high dimensionality of gene expression data. Often there are far more genes than subjects (p > n problem). The bclust library in R offers a Bayesian approach to unsupervised clustering and variable selection for high-dimensional data. In our research, we apply the bclust model to the cancer dataset to cluster the cancer types and assess which genes are most important in distinguishing the five cancer types.

This document is currently not available here.

Share

COinS
 
Nov 23rd, 8:00 AM Nov 23rd, 8:45 AM

Bayesian Clustering and Variable Selection for Gene Expression Data

247

Gene Expression data describes the activity level of genes in a particular cell type. All cells in an organism will consist of the same genes, however, depending on the cell type, the gene activity may vary. We consider a dataset describing the gene expressions of five different cancer types. We are interested in clustering the cancer types according to the various gene activity. Statistically, this is a challenge due to the high dimensionality of gene expression data. Often there are far more genes than subjects (p > n problem). The bclust library in R offers a Bayesian approach to unsupervised clustering and variable selection for high-dimensional data. In our research, we apply the bclust model to the cancer dataset to cluster the cancer types and assess which genes are most important in distinguishing the five cancer types.