Hi Wolfgang, I plot it on my screen in a x11 device. In that case it's extremely slow. But you can NOT distinguish the leaves. Of course. Creating this image through bmp(...) allows me to plot the tree in a much shorter time. Tony On 15 April 2010 00:14, Wolfgang Huber <whuber@embl.de> wrote: > > Hi Anthony > > just out of interest, what graphics device are you using to view the > heatmap with 8000 rows and the tree with 8000 branches? > > I wonder because if you will spend 1 mm per row, 8000 rows are 8 metres > [that's 4 centiinch and 26.2 feet in the US]. > > Best wishes > Wolfgang > > > Anthony Ferrari ha scritto: > >> Hi all, >> >> I have got a 8000*200 gene expression matrix. I am clustering rows and >> columns with hclust. >> >> I want to plot something like generated by 'heatmap' which means an >> heatmap >> with corresponding array and gene trees. So, the fact is that the gene >> tree >> should be horizontal. >> >> With 'plot' function applied to an hclust object, the dendrogram is always >> vertical. But plotting the tree is really fast. >> >> To plot the gene tree horizontal, I saw that coercing the hclust object >> with >> as.dendrogram allows one to use the option horiz=TRUE in 'plot'. That's >> pretty true but plotting the dendrogram for 8000 genes requires then an >> infinity of time. I see each branch of the tree be slowly drawn one by >> one. >> >> Why this difference in calculation time ? >> And is there a way to plot a horizontal tree as quick as >> 'plot(hclust_object,...)' does ? >> >> >> best regards, >> >> tony >> >> [[alternative HTML version deleted]] >> >> >> _______________________________________________ >> Bioconductor mailing list >> Bioconductor@stat.math.ethz.ch >> https://stat.ethz.ch/mailman/listinfo/bioconductor >> Search the archives: >> http://news.gmane.org/gmane.science.biology.informatics.conductor >> > > -- > > > Wolfgang Huber > EMBL > http://www.embl.de/research/units/genome_biology/huber > > > [[alternative HTML version deleted]]