Hi Syed,
Thanks for your reply:
Here is where you can fire the query to:
http://reactomedev.oicr.on.ca/cgi-bin/mart
And the canned query you want to try is "All modified proteins in a pathway".
The other canned queries are working fine.
Let me know if you need further details.
In the meantime, I will try your suggestions as well.
Many thanks,
Nelson

> Hi Nelson, not sure which server to post this query to. I tried
> reactome.org:5555 but no luck. I guess the second dataset isnt there.
> Anyhow, could you please do some detective work and start adding
> attributes of second dataset one by one and see which one causes it
> fail. Otherwise, let me have the URL of the server which has these
> datasets configured so i can fire this query.
>
> Best,
> Syed
>
> On 27/09/2011 16:35, [email protected] wrote:
>> Hi Syed,
>> Here is the XML equivalent of the query that works:
>>
>> http://pastebin.com/gcNrKnM8
>>
>> Thanks,
>> Nelson
>>
>>> Sorry Nelson, i might have missed few details, but could you send me
>>> the
>>> XML equivalent of the query ?
>>>
>>>
>>> On 27/09/2011 15:20, [email protected] wrote:
>>>> Hi Junjun&   Syed,
>>>> I still have problems with this earlier reported query. So I have
>>>> tried
>>>> to
>>>> tweak a few parameters (listed below) as I generated error logs,
>>>> hopefully
>>>> these will be more useful to diagnosing the problem.
>>>>
>>>> You correctly pointed out the query involves a join between two
>>>> datasets
>>>> linked by exportables and importables each of which has 2
>>>> attributes/filters (based on ensembl gene and uniprot id attributes).
>>>>
>>>> This query works with an old version of our database but it throws the
>>>> SQL
>>>> syntax error with the new version of the database.
>>>> I have tried to perform a few tests and saved the error_log files. I
>>>> hope
>>>> these will shed more light.
>>>>
>>>> You had mentioned that something might be missing before the 'AND' and
>>>> this reminded me that there possibly might be a NULL value on one of
>>>> the
>>>> tables that could trigger an ERROR (I have encountered this before,
>>>> and
>>>> I
>>>> got a patch from Damian on the FilterList.pm module to fix such a bug,
>>>> he
>>>> later committed this to the BioMart code). Since this mart is on a
>>>> different server from the one that had the patch, I thought I would
>>>> try
>>>> testing with the patch as well.
>>>>
>>>> Like Syed had asked, the query in question is sent to the server via
>>>> webservice.
>>>>
>>>>
>>>> Some information about the error_logs from each test:
>>>>
>>>> =================================================
>>>> These are logs for the query when its not working
>>>> =================================================
>>>>
>>>> About error_log1: http://pastebin.com/3CKRzTvc
>>>>    1. Deleted the two_variable_id exportable/importables both on EWAS
>>>> and
>>>> Pathway datasets.
>>>>    2. Used new version db (v38_mart).
>>>>    3. Query of interest not working.
>>>>    4. Used FilterList.pm module that was altered by Damian.
>>>>
>>>>
>>>> About error_log2: http://pastebin.com/ekVihAn9
>>>>    1. Deleted the two_variable_id exportable/importables both on EWAS
>>>> and
>>>> Pathway datasets.
>>>>    2. Used new version db (v38_mart).
>>>>    3. Query of interest not working.
>>>>    4. Used original FilterList.pm module.
>>>>
>>>>
>>>> About error_log3: http://pastebin.com/ZLagmDQh
>>>>    1. Has two_variable_id exportable/importables both on EWAS and
>>>> Pathway
>>>> datasets.
>>>>    2. Used new version db (v38_mart).
>>>>    3. Query of interest not working.
>>>>    4. Used FilterList.pm module that was altered by Damian.
>>>>
>>>> ==========================================
>>>> These are logs for the query when it works
>>>> ==========================================
>>>>
>>>>    About error_log4: http://pastebin.com/prAgP3vS
>>>>    1. Has two_variable_id exportable/importables both on EWAS and
>>>> Pathway
>>>> datasets.
>>>>    2. Used old version db (v35_mart).
>>>>    3. Query of interest Works.
>>>>    4. Used original FilterList.pm module.
>>>>
>>>>
>>>> About error_log5: http://pastebin.com/gS8HghkT
>>>>    1. Has two_variable_id exportable/importables both on EWAS and
>>>> Pathway
>>>> datasets.
>>>>    2. Used old version db (v35_mart).
>>>>    3. Query of interest Works.
>>>>    4. Used FilterList.pm module that was altered by Damian.
>>>>
>>>>
>>>>
>>>> Thanks,
>>>> Nelson
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>> This SQL doesn't look correct, something is missing right before the
>>>>> 'AND'
>>>>> for the part that is joined with 'OR', with the exception of the last
>>>>> item
>>>>> as below:
>>>>>
>>>>> ((pathway__referencepeptidesequence__dm.referencedatabase_uniprot =
>>>>> 'P49327') AND
>>>>> (pathway__referencednasequence__dm.referencedatabase_ensembl
>>>>> = 'ENSG00000169710'))
>>>>>
>>>>>
>>>>> It looks like the original query involves join between two datasets.
>>>>> Usually that is not a problem when the link defined as an
>>>>> exportable/importable pair, in which exportable contains one
>>>>> attribute
>>>>> (eg, ensembl_gene) and importable contains one filter (eg,
>>>>> ensembl_gene).
>>>>> The piece of SQL statement shown above makes me thinking that the
>>>>> link
>>>>> you
>>>>> defined seemed to have two attributes and two filters (ensembl gene
>>>>> and
>>>>> uniprot id) for its exportable and importable.
>>>>>
>>>>> Although it is legitimate to create such a link, but I am afraid it
>>>>> is
>>>>> not
>>>>> going to work in most of the cases. It works only when *one* pair of
>>>>> the
>>>>> values is exported from one dataset and imported to the other
>>>>> dataset.
>>>>> The
>>>>> 'band' filter works exactly like this in Ensembl gene mart, user
>>>>> specifies
>>>>> *one* band to a hidden dataset which exports *one* pair (ie, one row)
>>>>> of
>>>>> 'chromosome' and 'coordinate' value (eg, 1, 10000303), and
>>>>> ensembl_gene
>>>>> dataset imports this pair and stick it in to SQL that queries against
>>>>> it's
>>>>> own db.
>>>>>
>>>>> The confirm this, you can double check the exportable/importable
>>>>> settings
>>>>> in your configuration.
>>>>>
>>>>> Hope this helps. Let us know if you have any further questions.
>>>>>
>>>>> Best regards,
>>>>>
>>>>> Junjun
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 11-09-26 9:47 AM, "[email protected]"<[email protected]>   wrote:
>>>>>
>>>>>> Hi,
>>>>>> I am getting an error from one of my queries along the lines of SQL
>>>>>> syntax
>>>>>> being the cause, am not sure what the real problem is though. I set
>>>>>> the
>>>>>> log4perl.conf to debug mode and I got back a more elaborate error
>>>>>> report
>>>>>> http://pastebin.com/9rqKBfWR
>>>>>> Anyone know what the problem is?
>>>>>> Thanks,
>>>>>> Nelson
>>>>>>
>>>>>> Serious Error: Error during query execution: You have an error in
>>>>>> your
>>>>>> SQL
>>>>>> syntax; check the manual that corresponds to your MySQL server
>>>>>> version
>>>>>> for
>>>>>> the right syntax to use near 'AND
>>>>>> (pathway__referencednasequence__dm.referencedatabase_ensembl =
>>>>>> 'ENSG00000166' at line 1
>>>>>>
>>>>>> ERROR: caught BioMart::Exception::Database: Error during query
>>>>>> execution:
>>>>>> You have an error in your SQL syntax; check the manual that
>>>>>> corresponds
>>>>>> to
>>>>>> your MySQL server version for the right syntax to use near 'AND
>>>>>> (pathway__referencednasequence__dm.referencedatabase_ensembl =
>>>>>> 'ENSG00000166' at line 1
>>>>>>
>>>>>> _______________________________________________
>>>>>> Users mailing list
>>>>>> [email protected]
>>>>>> https://lists.biomart.org/mailman/listinfo/users
>>>>>
>>>>>
>>>>
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> Users mailing list
>>>> [email protected]
>>>> https://lists.biomart.org/mailman/listinfo/users
>>>
>>
>>
>


_______________________________________________
Users mailing list
[email protected]
https://lists.biomart.org/mailman/listinfo/users

Reply via email to