[BioC] optimizazion cluster in Heatmap

[Naomi Altman]

A possible problem is the choice of distance function.  The distance 
measure must match your idea of "similar expression profiles" - e.g. 
Euclidean distance emphasizes absolute expression, correlation 
distance emphasizes the same up and down pattern, etc. Centering and 
rescaling the gene expression values will change the distance, so you 
need to be careful about what you mean by "similar",

--Naomi

At 09:20 AM 11/8/2007, James W. MacDonald wrote:
>Hi Alessandro,
>
>I don't think anybody will be able to answer this question for you. If I
>use the same code on the sample.ExpressionSet that comes with the
>Biobase package, it seems to me that the rows are ordered quite well, so
>I wonder what exactly you don't like about the row ordering of your data.
>
>Anyway, there are any number of clustering methods you can use with the
>hclust() function, of which 'average' is only one. You might try some of
>the other clustering methods to see if you get something more to your
>tastes.
>
>Best,
>
>Jim
>
>
>
>Alessandro Fazio wrote:
> > Hello everybody, I have a problem in doing a cluster with Heatmap.
> > Briefly, I want a cluster with fixed column order and row order
> > depending on the dendrogram produced by the clustering method. BUT,
> > it seems that the row sorting is not optimal, that is genes with
> > similar expression profiles are not group tpgrther but spread. This
> > is the code I used: > mydist <- function(x) cor.dist(x)> myhclust <-
> > function(x) hclust(x, method='average')>
> > heatmap(exprs(exampleSet),Colv=NA, dist=mydist, hclust=myhclust) >
> > sessionInfo()R version 2.5.0 (2007-04-23) i386-pc-mingw32
> > locale:LC_COLLATE=English_United States.1252;LC_CTYPE=English_United
> > States.1252;LC_MONETARY=English_United
> > States.1252;LC_NUMERIC=C;LC_TIME=English_United States.1252attached
> > base packages:[1] 'splines'   'tools'     'stats'     'graphics'
> > 'grDevices' 'utils'    [7] 'datasets'  'methods'   'base'     other
> > attached packages:    bioDist          GO  genefilter    survival
> > ALL   Rgraphviz geneplotter     '1.8.0! '    '1.16.0'    '1.14.1'
> > '2.31'     '1.4.3'    '1.14.1'    '1.14.0'     lattice    annotate
> > Biobase        RBGL       graph    '0.15-4'    '1.14.1'    '1.14.0'
> > '1.12.0'    '1.14.2' ANY idea about what I should do to have a good
> > row sorting? THANK you in advance Regards, Alessandro
> > _________________________________________________________________
> > [[replacing trailing spam]]
> >
> > [[alternative HTML version deleted]]
> >
> > _______________________________________________ Bioconductor mailing
> > list Bioconductor at stat.math.ethz.ch
> > https://stat.ethz.ch/mailman/listinfo/bioconductor Search the
> > archives:
> > http://news.gmane.org/gmane.science.biology.informatics.conductor
>
>--
>James W. MacDonald, M.S.
>Biostatistician
>Affymetrix and cDNA Microarray Core
>University of Michigan Cancer Center
>1500 E. Medical Center Drive
>7410 CCGC
>Ann Arbor MI 48109
>734-647-5623
>
>_______________________________________________
>Bioconductor mailing list
>Bioconductor at stat.math.ethz.ch
>https://stat.ethz.ch/mailman/listinfo/bioconductor
>Search the archives: 
>http://news.gmane.org/gmane.science.biology.informatics.conductor

Naomi S. Altman                                814-865-3791 (voice)
Associate Professor
Dept. of Statistics                              814-863-7114 (fax)
Penn State University                         814-865-1348 (Statistics)
University Park, PA 16802-2111