Bug in 2.0.3?

4 views
Skip to first unread message

Wyatt VBI

unread,
Jun 24, 2011, 8:03:50 AM6/24/11
to Alternative Splicing and Functional Prediction
Hello everyone,

I'm new to AltAnalyze, but I am hooked!! This is one of those rare
things that is both easy to use and powerful. Thanks!

In the GO-Elite analysis of my most recent run of AltAnalyze, I got
some weird results. In my experiment:
There are 62,000 AS events in experimental condition "A".
There are 45,000 AS events in experimental condition "B".
There are 1200 AS vents in experimental condition "C".

This is largely what I would expect.
However, when I look at the GO analysis for the AS events in A, I only
get one ("1") significant GO term.
If instead I look at experiments B or C, I get 100s of GO terms.

Is it possible that AltAnalyze has mislabeled my results, putting GO-
Elite results from condition "B" into condition "C," and vice versa?
It only occurs in the AS results, the GE results correlate with what I
would expect.

Thanks,

Wyatt

Nathan Salomonis

unread,
Jun 24, 2011, 3:59:53 PM6/24/11
to alt_pre...@googlegroups.com, kwyatt....@ttuhsc.edu
Hi Wyatt,

Thank you for the positive comments. Although I can't think of any
reason this might happen, I haven't run many multi-condition anlayses
myself in AltAnalyze so it is worthwhile to explore. There are methods
I would suggest to find out what is going on:

Strategy 1:
Look in your results directory under the folders GO-Elite/input. Check
to see if the list of genes for each of the AltAnalyze comparisons
matches those in the folder AltResults/AlternativeOutput. The files
with the suffix "exon-inclusion-GENE-results.txt" should have the same
list of Ensembl genes. If not, this may be causing the problem. If
these look correct, check out the folder
GO-Elite/GO-Elite_results/CompleteResults/ORA/archived-date. This
folder contains statistics for all Gene Ontology and WikiPathway
terms. If one file has less terms than another, than the problem is
occurring here.

Strategy 2:
Run the files in the folder GO-Elite/input and GO-Elite/denominator,
in the GO-Elite web version (http://www.genmapp.org/go_elite/).
Assuming the list of genes is correct (corresponds to the correct
AltAnalyze results), these results should basically replicate those by
AltAnalyze. The difference comes with different versions of Ensembl
(the web version is still using Ensembl version 60 and not 62 yet). If
there is something strange about these results as well, just email me
the URL for the results folder.

Best,
Nathan

> --
> 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.
>
>

Mcmahon, Kwyatt

unread,
Jun 28, 2011, 6:28:01 AM6/28/11
to alt_pre...@googlegroups.com, nsalo...@gmail.com
Nathan,

Thanks for your response! I did what you suggested and it seems that there is nothing wrong with the program: I got the same results on the web! Who would have thought it - 60,000 AS events and only one over-represented GO term! That just wild!

Thanks for your help,

Wyatt
________________________________________
From: Nathan Salomonis [nsalo...@gmail.com]
Sent: Friday, June 24, 2011 2:59 PM
To: alt_pre...@googlegroups.com
Cc: Mcmahon, Kwyatt
Subject: Re: AltAnalyze User Group Bug in 2.0.3?

Reply all
Reply to author
Forward
0 new messages