Posttranslational modifications with multiple sites not valid

[ypriverol]

Hi @andrewrobertjones @timosachsenberg :

We have now the first submissions in mzTab. We have found that the specification is not clear when multiple sites can be associated with the same modification:

MTD     variable_mod[3]-site    N
MTD     variable_mod[3]-site    Q 

The previous example is failing because only one site can be associated with a modification. We can hard coded this by adding sites comma separated or we can accept the current version but we need to update the jmzTab.

We have multiple submissions in this case then feedback quick is more than welcome.

Yasset

[jgriss]

Hi @ypriverol,

All keys in the MTD section are defined to be unique. Therefore, the search settings you show should have been reported using two separate variable_mod objects.

I personally think that this is relatively clear in the specification document but we did not add a sentence stating that all keys MUST be unique.

Maybe we should add this as a clarification?

[ypriverol]

Then, for example for phosphorylation, we should add three variable modifications because we have three main possible sites (S/T/Y)

[timosachsenberg]

Yep. I agree with Johannes.

[jgriss]

Then, for example for phosphorylation, we should add three variable modifications because we have three main possible sites (S/T/Y)

Exactly. These are three independent objects.

[andrewrobertjones]

Agree - it's a bit clunky but it seems that each variable_mod_site combination has to be treated as a separate object

[ypriverol]

We need tom improve only the README. MaxQuant is now updating the export.

[ypriverol]

MaxQuant has been already updated. Thanks, @andrewrobertjones @timosachsenberg @jgriss for your comments.