I'm using this code to create a heatmap of expression data and indicate a
specific column of data, post clustering:
#Generate heatmap:
heatmapChipDx <- heatmap.2(L3data_med2,
dendrogram="column",trace="none",
col=redgreen(64), colsep=seps, sepcolor="yellow", sepwidth = 0.1,
margin=c(5,5), scale="row", key=FALSE)
I have two problems that I'm hoping i can get some pointers on:
1. The column I want to indicate with a yellow line is column #1 in the
matrix, pre-clustering. How can I determine its position post-clustering, so
I can instruct the heatmap code to move the yellow 'sep' line to this
position?
2. The intensity of the colours is very dull and I cannot figure out how to
increase the contrast (a la Treeview). I have attached two images (not sure
if this is allowed), one being the image as it comes out from the code
above, and the other ("...copy.png") being how I would actuallt like
it to
look. I adjusted the levels in photoshop to get this second image.
The data is log2 scale gene expression data (single channel), and scaled
with this command:
L3data_med<- scale(t(L3dataonly), center = TRUE)> L3data_med2<-t(L3data_med)
>
...to median (mean?) centre the rows, prior to clustering.
Thanks in advance,
Ryan
-------------- next part --------------
A non-text attachment was scrubbed...
Name: heatmap.png
Type: image/png
Size: 480026 bytes
Desc: not available
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20090615/d814291d/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: heatmap copy.png
Type: image/png
Size: 404978 bytes
Desc: not available
URL:
<https://stat.ethz.ch/pipermail/r-help/attachments/20090615/d814291d/attachment-0005.png>
