New PSICQUIC webservice

[Hardeep]

Hello,

I would like to announce a new webservice developed at the BAR (http://bar.utoronto.ca). The webservice queries a protein-protein interaction database which consists of more than 70944 predicted interactions and 28556 confirmed Arabidopsis interactions.

The webservice can be accessed here:

http://bar.utoronto.ca:9090/psicquic/webservices
and
http://bar.utoronto.ca:9090/psicquic/webservices/current/search/query/At1g10930?firstResult=0&maxResult=10

I would like to add the webservice to the PSICQUIC Registry. But before I do, I was wondering if I need to run it by someone in the PSICQUIC community to ensure I am using the correct PSICQUIC Ontology/format, or if there are any recommendations/improvements that can be done before PSICQUIC users start using the webservice.

thanks,

Hardeep
University of Toronto

[Lukasz Salwinski]

hmm... besides what looks like an overabundance of 'unknown' in the
example you point to above - what do you mean by 'protein protein
interaction' ? besides the columns providing gene/protein accessions,
taxon id and the one (I'm guessing confidence value ?) consistently
saying PCC:0.0 the server seems to provide no information about
interaction type and experimental method used to infer it ?

lukasz


--
-------------------------------------------------------------------------
Lukasz Salwinski PHONE: 310-825-1402
UCLA-DOE Institute for Genomics & Proteomics FAX: 310-206-3914
UCLA, Los Angeles EMAIL: luk...@mbi.ucla.edu
-------------------------------------------------------------------------

[Rafael Jimenez]

Hi Hardeep,

Great job! ... sure the community will value to have a new PSICQUIC service providing a comprehensive source of molecular interactions for Arabidopsis.

As Lukasz suggested there are some things might need clarification or improvement. A good place to start is the Data Distribution Best Practises guideline: http://code.google.com/p/psicquic/wiki/DataDistributionBestPractices

I would recommend you to make changes to your service following this guideline. Afterwards I will be very happy to have a look to your PSICQUIC service and include it in the registry.

Best regards,
Rafael.

--
You received this message because you are subscribed to the Google
Groups PSICQUIC group.
To post to this group, send email to psic...@googlegroups.com
To unsubscribe from this group, send email to
psicquic+unsubscribe@googlegroups.com

http://psicquic.googlecode.com

[Hardeep]

Hi Lukasz and Rafael,

Thanks for the recommendations. I'm happy to address the suggestions you mentioned. I'll keep you posted.

best,
Hardeep

psicquic+u...@googlegroups.com

http://psicquic.googlecode.com

[harde...@yahoo.ca]

Hi Rafael, hi Lukasz,

I've implemented your suggestions and included information about the experimental method and interaction type for our PSICQUIC webservice. Here is an example:

http://bar.utoronto.ca:9090/psicquic/webservices/current/search/query/At2g26990 NOT "split ubiquitin"

There are a few interactions for which the experimental method/interaction type is not defined but we plan to add this information soon. Rafeal - I have a question regarding the "source". Since our protein-protein interaction data is aggregated from various interaction databases, should we list the source as the name of the database from which the interaction came from (ie. some of the interactions are from the Mind database at http://cas-biodb.cas.unt.edu/project/mind/index.php), or the name of our database (BAR)? We have included the appropriate publications for each interaction in such cases. I look forward to your feedback.

thanks,
Hardeep

[Rafael Jimenez]

Hi Hardeep,

On 19 December 2012 22:32, <harde...@yahoo.ca> wrote:

Hi Rafael, hi Lukasz,

I've implemented your suggestions and included information about the experimental method and interaction type for our PSICQUIC webservice. Here is an example:

http://bar.utoronto.ca:9090/psicquic/webservices/current/search/query/At2g26990 NOT "split ubiquitin"

Good job!
 

There are a few interactions for which the experimental method/interaction type is not defined but we plan to add this information soon. Rafeal - I have a question regarding the "source". Since our protein-protein interaction data is aggregated from various interaction databases, should we list the source as the name of the database from which the interaction came from (ie. some of the interactions are from the Mind database at http://cas-biodb.cas.unt.edu/project/mind/index.php), or the name of our database (BAR)? We have included the appropriate publications for each interaction in such cases. I look forward to your feedback.

Ideally you should mention the original sources plus the source reporting the interaction, in your example it would be "Mind" and "Bar". This is better described in PSI-MITAB 2.7  ...

13. Source databases and identifiers, taken from the corresponding PSI-MI controlled vocabulary, and represented as databaseName:identifier(sourceName). As the detection methods are taken from the PSI-MI ontology, the database name is 'psi-mi'. Multiple source databases can be separated by "|". When the interaction has been imported and reported by different sources, it is recommended to give the original source plus the source that currently reports the interaction. Ex: psi-mi:"MI:0469"(intact)|psi-mi:"MI:0923"(irefindex)

"Mind" and "Bar" seems to not be present in the ontology. I will make a request to add them.

I am happy to have a look to your service in more detail, but I will not be able to do it until I am back from holidays. Would be OK if I work on it as soon as I am back around the the 2nd of January?

Best regards,
Rafael.