mapping file mayhem

[fgrunewald]

When a modification is part of an atomistic base force-field, it needs to have mapping files in ALL other force-fields, which it can map to. If that is not the case the modification is not skipped but causes an error. So at some point all martini force-fields have to be homogenized.

[pckroon]

There should only be an error if you try to use that modification, no? Modifications/blocks being unmapped to some CG FF shouldn't be a problem until you try to use them.

[fgrunewald]

the problem is that martinize2 will automatically try to apply the modification if it finds it in the input. For example, we have the protonated histidine modification present in the AA files. Now for ALL martini force-fields it will try to apply the modification and then fail because the corresponding files are either not present or incorrect

[pckroon]

This is desired behaviour, since it will result in a warning, not an error. And that warning is important and informative.

EDIT: That said, it's definitely a good idea to add all those modifications to the other CG FFs

[fgrunewald]

It appears though as if the modifications, which are old and not connected components at the CG level, will now cause an error instead of a warning.

[pckroon]

Oh that could be, indeed. That needs to be fixed.