[BioC] heatmap genenames

[michael watson (IAH-C)]

Hi Claire

I will have a look at this tomorrow when I have R actually in front of me, but in the meantime, may I reccomend that you specifically use hclust and dist when using heatmap?  I always do - it gives you more control over what exactly is going on!

So do:

d <- dist(mydata)
hc <- hclust(d)
heatmap(mydata, RowV=as.dendrogram(hc))
gene.order = hc$order

Mick


-----Original Message-----
From:	Claire Wilson [mailto:ClaireWilson at picr.man.ac.uk]
Sent:	Thu 11/4/2004 5:03 PM
To:	alessandro ambrosi
Cc:	BioC mailing list
Subject:	RE: [BioC] heatmap genenames

Hello,

Many thanks for your reply, I have tried repeating the clustering done
by heatmap using hclust to obtain the row order as follows:

d <- dist(mydata)
hc.d <- hclust(d)
gene.order <- hc.d$order

and compared the order to what I see on a heatmap when I do

heatmap(mydata)

However, it would appear that the ordering of genes within a heatmap
made from heatmap(mydata) is different to that seen if a dendrogram is
made from from hc.d. So I am thinking something else happens after doing
the hierarchical clustering within the heatmap function, but I have so
far been unsuccessful in my attempts to try and work out what it is.

Claire


> Dear Claire,
> 
> if I remember well heatmap uses hclust as default clustering 
> function with
> distances given by "dist" that, in turn, uses "euclidean" 
> distance as default.
> 
> So, I guess that you should be able to obtain the desired 
> order easily runing
> 
> hclust(dist( your.data ))$order
> 
> or something like
> 
> my.cluster<-hclust(dist( my.data ))
> rownames( my.data )[my.cluster$order]
> 
> 
> Let me know if it works
> Alessandro
> 
> 
> At 15.06 04/11/2004, you wrote:
> >Dear all,
> >
> >I have been using the heatmap function to visualise my data and would
> >like to obtain a list of my genes in the same order that 
> they appear in
> >the heatmap. I have looked through the mail archives and tried just
> >creating a dendogram and using order.dendogram but have so far been
> >unsuccessful. Can anyone point me in the right direction?
> >
> >cheers
> >
> >claire
> >
> >--------------------------------------------------------
> >
> >
> >This email is confidential and intended solely for the use 
> o...{{dropped}}
> >
> >_______________________________________________
> >Bioconductor mailing list
> >Bioconductor at stat.math.ethz.ch
> >https://stat.ethz.ch/mailman/listinfo/bioconductor
> 
> 
> 
>
 
--------------------------------------------------------

 
This email is confidential and intended solely for the use o...{{dropped}}

_______________________________________________
Bioconductor mailing list
Bioconductor at stat.math.ethz.ch
https://stat.ethz.ch/mailman/listinfo/bioconductor