Alt analyze question

19 views
Skip to first unread message

Eugene Bolotin, PhD

unread,
Sep 21, 2011, 1:41:50 PM9/21/11
to alt_pre...@googlegroups.com
Hello,
I have a dataset from illumina bead arrays, unfortunately it is in a giant text file with about 1000 columns (experiments) and a row for each probeset. I analyzed it using limma R packge. I know the illumina array is not designed for this, but I would like to get alternate splcing data on genes with multiple probes in this array. Is there a way to do it with altanalyze? Can you reccomend any other softwares for the task?
Thanks,
Eugene

--
Eugene Bolotin, Ph.D.
Genetics Genomics and Bioinformatics
CHORI (Children's Hospital Oakland Research Institute)
5700 Martin Luther King jr Way
Oakland, CA 94609

Nathan Salomonis

unread,
Sep 21, 2011, 2:51:04 PM9/21/11
to alt_pre...@googlegroups.com
Hi Eugene,

This is now possible in the current version of AltAnalyze. I just
posted a page outlining how to do this on our wiki:
http://code.google.com/p/altanalyze/wiki/SplicingForAnyPlatform

The key is to reformat your data into a universally generic import
format (UCSC BED) and treat this data as if it is RNA-Seq expression
data. Instead of counts, you may have intensity data, which is fine. A
1000 columns is fine, but you will just want to make sure that you
create your groups and comparison file before running AltAnalyze as
described here:
http://code.google.com/p/altanalyze/wiki/ManualGroupsCompsCreation

I have not worked with non-Affy and non-RNA-Seq data yet, but there is
no reason I know of why this should work. Please contact me if you are
having problems with this.

Best,
Nathan Salomonis

> --
> You received this message because you are subscribed to the Google Groups
> "Alternative Splicing and Functional Prediction" group.
> To post to this group, send email to alt_pre...@googlegroups.com.
> To unsubscribe from this group, send email to
> alt_predictio...@googlegroups.com.
> For more options, visit this group at
> http://groups.google.com/group/alt_predictions?hl=en.
>

Reply all
Reply to author
Forward
0 new messages