when is it advantageous to use automatic linking over supplying a .params file with the link parameters defined to refine links? I'm sure Nigel will reply with the best answer. What I can say is that automatic linking links atoms based on a variety of guess-work and some (mostly arbitrary) cutoffs (like, for instance, if two atoms are close within 3A let's postulate they are covalently

Hi Georg, linked, etc etc). What you specify in parameter file is the instruction for the program to do exactly that: if the parameter file says there is a link between two atoms, the program must obey and create that link. If it does not, please report a bug.

I thought that automatic linking will recognize the kind of link by itself and then look for the restraints itself. But it seems not to do, because it is not complaining that there are still hydrogens in the sugar moeity which should not be there after the link has formed.

This brings me to my second questions. Is there a way to automatically remove the hydrogens which should not be there anymore after the bond has formed. Or tell reduce that it should not add hydrogens at the link?

I might be wrong (Nigel to correct me), but I doubt H are removed automatically when necessary to form a link. So I would use inputs that are as close as possible to what you want, and also instruct the program about what you want, and then hope it does what you want. Again, if it doesn't, please report a bug. Good luck! Pavel

Hi Ben, actually the algorithm for the automatic hydrogen addition doesn't take links into account. You could add the hydrogens in an extra step (phenix.reduce) before the refinement and remove afterwards the few ones which are misplaced in a sugar link. I do it in that way, because the refinement with hydrogens lead to a better overall geometry and at the end better statistics. Christian Am 18.06.2014 20:14, schrieb Benjamin Stauch:

All,

I believe I had the same issue a couple of days ago (check the mailing list archive for reference ..) -- I was introducing a Asn-NAG-NAG albeit not using automatic but explicit linking (I can provide the necessary file if needed) and I ran into trouble with hydrogens -- Phenix was deleting them after linking, however b/c I had checked the "automatically add hydrogens" option in the GUI it was giving me an error for unparametrised hydrogens. So I had to make sure to UNcheck the add hydrogens box.

HTH

Ben

2014-06-18 8:08 GMT+02:00 Pavel Afonine mailto:[email protected]>:

Hi Georg,

...

when is it advantageous to use automatic linking over supplying a .params file with the link parameters defined to refine links?

I'm sure Nigel will reply with the best answer. What I can say is that automatic linking links atoms based on a variety of guess-work and some (mostly arbitrary) cutoffs (like, for instance, if two atoms are close within 3A let's postulate they are covalently linked, etc etc). What you specify in parameter file is the instruction for the program to do exactly that: if the parameter file says there is a link between two atoms, the program must obey and create that link. If it does not, please report a bug.

...

I thought that automatic linking will recognize the kind of link by itself and then look for the restraints itself. But it seems not to do, because it is not complaining that there are still hydrogens in the sugar moeity which should not be there after the link has formed.

This brings me to my second questions. Is there a way to automatically remove the hydrogens which should not be there anymore after the bond has formed. Or tell reduce that it should not add hydrogens at the link?

I might be wrong (Nigel to correct me), but I doubt H are removed automatically when necessary to form a link. So I would use inputs that are as close as possible to what you want, and also instruct the program about what you want, and then hope it does what you want. Again, if it doesn't, please report a bug.

Good luck! Pavel

_______________________________________________ phenixbb mailing list [email protected] mailto:[email protected] http://phenix-online.org/mailman/listinfo/phenixbb

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb