Question

Querying clustering in heatmap.2

0

Entering edit mode

Elizabeth Ashley ▴ 50

@elizabeth-ashley-4202

Last seen 10.2 years ago

Hi I have used heatmap.2 to create a heatmap of my data. It automatically clusters the data and I can't identify which part of the text does this. However I am quite happy to have the clustering (of GO categories) but would like to be able to query the groupings to find insights into how it is clustered. Are these clusters named, and if so can you find out which ones bring back significant data? Below is the script I am using. GO<-read.delim("file.txt",header=TRUE) GO GO[,-c(1)] row.names(GO)<-GO$NAMES source("http://www.bioconductor.org/biocLite.R") biocLite("ALL") library("Ringo") GO_matrix<-data.matrix(GO[,-c(1)]) source("http://www.bioconductor.org/biocLite.R") biocLite("limma") library(limma) library(gplots) heatmap.2(GO_matrix,col=redgreen(256), margins=c(5,10),xlab="DATASET",ylab="GO TERM", key=TRUE,keysize=1,symkey=FALSE,density.info="none", trace="none",cexCol=0.5,cexRow=0.1) Thanks for any help you can provide with this matter Lizzy

Clustering Clustering • 1.2k views

ADD COMMENT • link updated 14.2 years ago by Steve Lianoglou ★ 13k • written 14.2 years ago by Elizabeth Ashley ▴ 50

score 0 · Answer 1 · 2010-09-15

Hi Lizzy, On Wed, Sep 15, 2010 at 7:59 AM, Elizabeth Ashley <elizabeth.ashley at="" linacre.ox.ac.uk=""> wrote: > Hi > > I have used heatmap.2 to create a heatmap of my data. It automatically clusters the data and I can't identify which part of the text does this. However I am quite happy to have the clustering (of GO categories) but would like to be able to query the groupings to find insights into how it is clustered. Are these clusters named, and if so can you find out which ones bring back significant data? Below is the script I am using. I don't really know what type of data is in your GO matrix, nor do I understand how you intend to find out "which ones bring back significant data" but an answer to your more general question is to look at the help from the heatmap.2 function: "?heatmap.2" Scroll down to the "Value" section, and you'll see that there is a list/structure that is returned "invisibly" (see ?invisible). There are two elements in the list named "rowInd" and "colInd" as well as "rowDendogram" and "colDendrogram" which you can use to start investigating which rows + cols are grouping together. Hope that helps, -steve -- Steve Lianoglou Graduate Student: Computational Systems Biology ?| Memorial Sloan-Kettering Cancer Center ?| Weill Medical College of Cornell University Contact Info: http://cbio.mskcc.org/~lianos/contact