search

vloux / ProteoRE

GNU General Public License v3.0
2 stars 5 forks source link

New tool: Add_Protein_Features (Rattus) - uniprot #263

Closed yvandenb closed 4 years ago

yvandenb commented 4 years ago

Same specifications as "Add protein features (Mouse) [UniProt]" With Taxonomy ID: 10116 for Rattus norvegicus

combesf commented 4 years ago

Done for Rat and Human (I thought you would ...) To be tested on migale-proteore, tool "Add protein features (3 orga) [UniProt]" (1st tool listed in the tool panel)

yvandenb commented 4 years ago

History "Test_Idconverter" shared with Flo, David, Val Run using two ID lists rattus and homo sap => execution failed image

combesf commented 4 years ago

Done for Rat and Human (I thought you would ...) To be tested on migale-proteore, tool "Add protein features (3 orga) [UniProt]" (1st tool listed in the tool panel)

History "Test_Idconverter" shared with Flo, David, Val Run using two ID lists rattus and homo sap => execution failed image

I looked at the pb. 1st : one protein features was missing and was unable to be retrieved : it is coorected now. So it runs and gives a result.

but 2d : for your rats ID, none of them match the proteome file from Uniprot ... I downloaded the reviewed and unreviewed proteome from rat and human. @davidchristiany did you donwload the same, or only the reviewed one ? I did not find the info thanks.

yvandenb commented 4 years ago

Ok thx Flo

for your rats ID, none of them match the proteome file from Uniprot . This situation should not raised a "Fatal error", instead should be handled by the tool; for instance by notifying explicitly the end-user in the history item e.g. "IDs not found" For your info, I submitted the Rattus ID list to Uniprot and all of these entries have a "reviewed" status...

combesf commented 4 years ago

Indeed the "fatal error" was due to the 1st pb, which is fixed now. No "fatal error" now if you run again

yvandenb commented 4 years ago

I 've been running again the newer version of the tool with two human IDs list => fatal error again image Don't know why Please could you check on your side ? Minor change: please make the name of the output in agreement with the name of the species selected (presently Mouse whereas Human was selected)...

combesf commented 4 years ago

You ran the Galaxy Version 2019.12.13.2, which is nit the latest version. The lastest version is the 2019.12.16.3 in which the minor change you mentioned has already been fixed. ("mouse" is not mentionned anymore)

yvandenb commented 4 years ago

Sorry :-/ now it's ok...

combesf commented 4 years ago

I checked: on the reviewed rat proteome (not the version on the online tool) , your rat IDs are there. Would you want me to replace the unreviewed proteomes (human, mouse and rat) by the reviewed ones ?

yvandenb commented 4 years ago

Not sure I understood your comment...normally both types of ID (reviewed and unreviewed) are in Uniprot ? How features are actually retrieved from Uniprot by this tool ?

combesf commented 4 years ago

I'll explain you de visu. I just updated the proteomes files (version 2019.12.16.4) and I think results will be better ...

combesf commented 4 years ago

As discussed files have been updated : now it is reviewed + unreviewed. To be tested, than to be installed on proteore.org ?

combesf commented 4 years ago

PS : to be tested on migale-proteore

combesf commented 4 years ago

@yvandenb should I close this issue ?

yvandenb commented 4 years ago

Works in my hands..please add in the UserDoc section => "Data source (release date)" "Annotations have been retrieved from UniProt.org (December 2019)." the url of the reference proteomes and exact release/number/version used alos complete the "http://www.proteore.org/static/data_source.html" file accordingly