Hi Christoph,
> Wim Vranken schrieb:
> > Explicitly choosing resonances is not really a good option, as you would
> > quickly end up with over 1000 for a normal-sized protein.
> >
> oops. got to be more precise with my wording. I meant the shift list
> that is created when importing the .prot file. Just consider what would
> happen if you try to import a second .prot .peaks pair with a different
> atom->number mapping for the same protein.
It will always use the information from the last imported .prot file...
this is not perfect, but works in almost all cases. The case where it will
fall over is when you import a .prot/.peak pair for one protein, and then
a .peak file for another protein without importing the correct .prot file
first. Another problem could arise if you import .prot/.peak pairs for
different proteins without running linkResonances in between. If you link
after importing the information for each protein there shouldn't be any
problems.
That said, if you do have this situation and have to redo your import
multiple times, it is easy enough to write a (Python-level) script that
does everything for you exactly the way you want it. Things are much more
customisable on the Python level... you can set more options and do
everything in the correct order.
As yet you cannot tell linkResonances exactly which resonances to link (if
they're a subset of the currently unlinked resonances), but I can build
that in if someone needs it.
Wim
|