Can you please be specific about the problems that you're having and the datasets that exhibit these problems?

Thanks,
J.

On May 11, 2011, at 10:51 AM, shamsher jagat wrote:

I have shared the history with you Jeremy and has named the files accordingly. I still have same problem and file can not be uploaded to UCSC genome browser
by clicking icon of browser on Galaxy.
Thanks
On Mon, May 9, 2011 at 6:21 AM, Jeremy Goecks <jeremy.goe...@emory.edu> wrote:
(Starting new thread on galaxy-user.)

Jagat,

It depends what filter tool you're using and what dataset you're filtering. There is a generic filter tool that can be used to filter Cuffdiff tabular files for either FPKM values and differential expression tests. There is also a tool for filtering GTF files based on a Cuffdiff expr dataset. It sounds like you may be confusing either the tools or the inputs.

If after double-checking you're still having problems with filtering, please put together a short list of your analysis steps and share your history with me, and I can take a look.

Thanks,
J.

Further to my question, It appear that there is some problem with the filter option: When I use the isoform/gene exp file as such it work fine but when I filter these files with either parameter such as status if test was successful or on p value it return me empty file. The way am saving the file is - expr file filter save as txt file and upload back in Galaxy.
Any suggestion?

Jagat


On Tue, May 3, 2011 at 3:08 AM, shamsher jagat <kanwar...@gmail.com> wrote:
Jeremy,

I have been trying to follow the steps in filtering Cufflink out put files you have described in one of the previous messages (http://gmod.827538.n3.nabble.com/Re-downstream-analysis-of-cuffdiff-out-put-td2836457.html ):

I have shared histroy with you, but in summary:

File 35: when Filter GTF data by attributes value list on data 11 (combined GTF) and data 33 (which is gene expr file) . Will not this should have one gene per row. But it is not?

File 39: Filter GTF file by attribute value list on data 11 and data 38 (Cuffdiff splicing expr) it failed. I would assume that it should filter on the basis of TSSid . The error message is

Traceback (most recent call last):
File "/var/opt/galaxy/g2test/galaxy_test/tools/filters/gff/ gtf_filter_by_attribute_values_list.py", line 67, in
    filter( gff_file, attribute_name, ids_file, output_file )
File "/var/opt/galaxy/g2test/galaxy_test/tools/filters/gff/ gtf_filter_by_attribute_values_list.py", line 57, in filter
    if attributes[ attribute_name ] in ids_dict:
KeyError: 'tss_id'

40 : Filter GTF data by attribute list on data 11 and 34 (tss group exp) failed and error message is:

Traceback (most recent call last):
File "/var/opt/galaxy/g2test/galaxy_test/tools/filters/gff/ gtf_filter_by_attribute_values_list.py", line 67, in
    filter( gff_file, attribute_name, ids_file, output_file )
File "/var/opt/galaxy/g2test/galaxy_test/tools/filters/gff/ gtf_filter_by_attribute_values_list.py", line 57, in filter
    if attributes[ attribute_name ] in ids_dict:
KeyError: 'tss_id'

I would consider that if one gene has different Id than there is splicing .

However in contrast isoform file with transcript Id is working fine (File 20)

On a different note can I convert GTF file to txt tab delaminated file I tried to convert file 11 in txt (following Edit attributes) but the file is not properly formatted especially col-pid and TSS id. Am I doing something wrong.

Thanks.



___________________________________________________________
Please keep all replies on the list by using "reply all"
in your mail client.  To manage your subscriptions to this
and other Galaxy lists, please use the interface at:

 http://lists.bx.psu.edu/



___________________________________________________________
The Galaxy User list should be used for the discussion of
Galaxy analysis and other features on the public server
at usegalaxy.org.  Please keep all replies on the list by
using "reply all" in your mail client.  For discussion of
local Galaxy instances and the Galaxy source code, please
use the Galaxy Development list:

  http://lists.bx.psu.edu/listinfo/galaxy-dev

To manage your subscriptions to this and other Galaxy lists,
please use the interface at:

  http://lists.bx.psu.edu/

Reply via email to