Any problems in RNA-seq anlysis

[GoogleCodeExporter]

AltAnalyze version 2.01 alpha_Py was run with default parameters and the sample 
data "hESC-NP_BED_files.zip", we get any error "print 
len(junction_db),'junctions present in',algorithm,'format BED files.' # 
('+str(pos_count),str(neg_count)+' by strand).'
UnboundLocalError: local variable 'algorithm' referenced before assignment";

thanks for your attention!

Original issue reported on code.google.com by fen...@aliyun.com on 22 Feb 2011 at 7:42

[GoogleCodeExporter]

Oddly, I don't get this error in version 2.01 alpha, although it does crash due 
to a memory error with the sample data due to one step in the program that is 
now removed (intron only to intron only junction comparisons). I just tested 
the sample data with the latest version (2.02 beta) and it works fine on 
Windows 64bit. I know this error occurred in Linux so I will see if I can 
reproduce it there. Try the latest version out and see how it goes.

Original comment by nsalomo...@gmail.com on 28 Feb 2011 at 7:04

[GoogleCodeExporter]

thanks for your timing, I have resolved this problem with version 2.02 beta. 
Thanks to Nathan Salomonis.

Original comment by fen...@aliyun.com on 28 Feb 2011 at 2:12

[GoogleCodeExporter]

I am using version 2.0.7 for Ubuntu. Things work really well until I get what 
seems like a Python related error when statistics get calculated:

15166 IDs imported...beginning to calculate statistics for all group comparisons
Traceback (most recent call last):
  File "<string>", line 4916, in AltAnalyzeSetup
  File "<string>", line 4374, in __init__
  File "<string>", line 5207, in AltAnalyzeMain
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/ExpressionBuilder", line 1359, in remoteExpressionBuilder
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/ExpressionBuilder", line 145, in calculate_expression_measures
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/reorder_arrays", line 271, in reorder
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/statistics", line 161, in moderateTestStats
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/statistics", line 416, in getModeratedStandardDeviation
  File "/home/nsalomonis/Downloads/pyinstaller-pyinstaller-b8ea399/AltAnalyze/build/pyi.linux2/AltAnalyze/out03-PYZ.pyz/statistics", line 429, in NewtonInteration
NameError: global name 'self' is not defined

I have attached the full .log file for further debugging. 

Otherwise, this looks like a fantastic piece of software. Thank you!

Original comment by vincent....@gmail.com on 8 Feb 2013 at 9:21

Attachments:

• AltAnalyze_report-20130208-145537.log

[GoogleCodeExporter]

I should add that I've done all the Python module installations that are 
mentioned in the Google code wiki. Apparently there's an issue with the 
NewtonIteration method but it's hard to tell what's up (at least for a 
bioinformatics newbie like myself). 

Original comment by vincent....@gmail.com on 8 Feb 2013 at 9:23

[GoogleCodeExporter]

Hi Vincent,

This appears to be related to a bug that was recently fixed in our unreleased 
dev version, but not in the version online. This will not occur for all 
datasets analyzed but just those which require an extra step in the moderated 
test calculation.

If running by command-line, add: --groupStat 'unpaired t-test'
This will use the un-moderated standard t-test instead.

If running using the GUI, select the unpaired t-test option in the Expression 
Analysis Parameters menu rather than the moderated. 

Since the initial chromosomal coordinate alginment step can be long, you can 
start from the all ready processed exp.YourExp.txt (not 
exp.YourExp.steady-steady.txt) and it will work fine (see documentation for 
--expdir option or Expression File in the GUI). Also, if using the Ubuntu 
compiled version you shouldn't need any python dependencies (will be included). 
Sorry about the issue.

Original comment by nsalomo...@gmail.com on 8 Feb 2013 at 10:51

[GoogleCodeExporter]

Thanks! That worked perfectly.

One more question: My workstation at the research center here has a proxy. I 
have established the right proxy permissions for the Terminal and for python in 
general (export http_proxy="http://myproxy.edu:3128" and that sort of thing), 
but can't seem to get AltAnalyze to download things like the 
LineageProfiler.zip and other files from the web directly. Is there an easy way 
to configure AltAnalyze so as to give it proxy permission?

Original comment by vincent....@gmail.com on 13 Feb 2013 at 8:58

[GoogleCodeExporter]

Hi,

I could not choose RNA-seq as my method of analysis for the species I am 
working on. The only option is Affymetrix microarray. What do I do? (All I have 
is an excel spread sheet of filtered RNA-seq reads as RPKM from the 
collaborator.)

Thanks, 

Original comment by wum...@gmail.com on 30 Nov 2013 at 4:46