hclust failed with very large (?) data

Question

[mbk0asis]

Hello

I was trying to draw a heatmap with hclust + heatmap.2, but it failed.

I think it's because my data is too large.

My data has only 2 columns but has more than 130,000 rows.

Is there a way to avoid it?

Thank you!

Answer 1

[samuelmiver]

If this is a gene expression array dataset, most of the array features are not going to change across treatments/conditions and will be uninformative.

I would try using a filter which does not use treatment/condition information in order to decrease the array dimension. There are some examples in the affycoretools and genefilter packages from R Bioconductor http://www.bioconductor.org/

[mbk0asis]

Thank you, samuelmiver, for the comment.

I will try to remove uninformative regions.