Rappsilber-Laboratory / xiview

GNU General Public License v2.0
4 stars 1 forks source link

Manual addition of mod masses in XiView -> use default from Xi01?! #444

Open ludsinn opened 5 years ago

ludsinn commented 5 years ago

Hi,

when I submitted my search results to XiView, I was asked to enter the masses of the modifications (btw, which one: average or monoisotopic?) which are were all set in Xi01 and are kind of standard in the field. An automatic proposal would be nice to have in there ;-)

Cheers Ludwig

colin-combe commented 5 years ago

The next time you upload to xiView it will auto-propose the values you used the last time (assuming modification names are the same).

If you want to completely avoid this, upload mzIdentML which already contains the modification mass info.

which one: average or monoisotopic?

er, good point, I'm not sure. The dialogue should say which. @lars-kolbowski - which is it, Lars?

which were all set in Xi01 and are kind of standard in the field.

It seems to be less standard than you might think (e.g. the current xi database has multiple modifications with same name but different mod. mass).

But the real problem is there's no standard what-so-ever for the names. So, even if there are standard values for modification masses we don't know which one the short name refers too. Speak to Lars about it, he's the official Global Ambassador for Universal Short Modification Names.

ludsinn commented 5 years ago

Thank you, Colin!

According to Lutz, it is always monoisotopic mass one should use...

I will hunt the ambassador to get the naming issue solved ;-)

colin-combe commented 5 years ago

As I understand it, it is this project: https://github.com/HUPO-PSI/psi-mod-CV (more specifically this file: https://raw.githubusercontent.com/HUPO-PSI/psi-mod-CV/master/PSI-MOD.obo ) where standardised names for modifications would be helpful.

I know Lars is super excited about his role in this ;) Just kidding, probably there will never be standardised names and the real solution is using mzIdentML.