A comparison tree for multiple RNA-seqs

[allenma]

Is there a program that can compare 20 RNA-seqs (with different treatments in the same cell line) to one another and determine which samples are the most similar and which are most different? It would be kind of like a phylogenetic comparison... or an RNA-seq BLAST.

I'm not sure what his would even be called so searching for it is hard.

Thanks

[allenma]

BTW, I don't want to do a gene expression comparison but a whole sample to whole sample.

[kopi-o]

Do you mean a hierarchical clustering? You could do a heatmap with hierarchical clustering easily in R, for instance.

[pbluescript]

R has a ton of packages for this ranging from simple regression models to more complicated machine learning algorithms for clustering. What type of comparison do you want to do?
If you're not familiar with R, you could give this program a shot:

http://www.exploredata.net/

Here's the paper describing it:

http://www.sciencemag.org/content/334/6062/1518.abstract?sid=0c9da855-2f28-461a-bb8e-a48b1f07cfa7

[allenma]

Thanks for these answers. I will try them.

[Simon Anders]

Have a look at the section on sample clustering and PCA in the new DESeq vignette.

[ekimmike]

what would you recommend for plotting heatmaps of multiple conditions?
I really appreciate DEseq for pairwise comparison but having more than 10 conditions make analysis a bit problematic especially when it comes to heatmaps, and comparison between samples

[upendra_35]

How about doing MDS plot? It can be easily done in R.

[ekimmike]

not enough I was asked for heatmaps, to reveal nice clusters and so...
this is not that straightforward having multiple conditions in DEseq analysis