We are currently working on a microarray project where we have to correlate both clinical and morphological (MRI) data with human expression data. Much of the data is in the form of continuous variables. While it can be turned into the categorical classifications most often used in array analysis, we would also like to look for correlations between the continuous variables and gene expression patterns. Does anyone on this list know of any good references on this subject? Also are there any existing Bioconductor packages which support this kind of analysis. thanks David Pritchard Research Scientist Dept. of Pathology University of Washington

An embedded and charset-unspecified text was scrubbed... Name: not available Url: https://stat.ethz.ch/pipermail/bioconductor/attachments/20060813/ 5ad14642/attachment.pl

On 8/13/06 9:58 PM, "David K Pritchard" <dpritch at="" u.washington.edu=""> wrote: > We are currently working on a microarray project where we have to > correlate both clinical and morphological (MRI) data with human expression > data. Much of the data is in the form of continuous variables. While it can > be turned into the categorical classifications most often used in array > analysis, we would also like to look for correlations between the continuous > variables and gene expression patterns. > Does anyone on this list know of any good references on this subject? > Also are there any existing Bioconductor packages which support this kind of > analysis. You might want to look at the possibility of applying random forests. They can be used for both classification and regression. Sean