-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 Hi Ed,
I had assumed that you had not yet refined with dual conformation but only one and see one extra peak at the position of the second CE. In that case your map would be biased by having only one conformation. Even if the coordinates of the remaining atoms were identical within the precision of the PDB file I would still split the full side chain, because: - - you don't know what is going to happen all through the end of your refinement - - in general, rather than your specific case, I consider it bad practice to not split the full side chain but restrict the multiple conformation to a single atom. Maybe this is the reason coot does not offer this option. You can still do it with a single line edit of the PDB file and adding 'A' to the first CE. Best, Tim On 06/07/2012 05:43 AM, Ed Pozharski wrote: > Tim, > > that is of course different from my original question, which was > more about coot capabilities. > > It's ~1.7A, and I am not sure what you mean by model bias. To be > more specific, there is a positive density peak suggesting > alternative position for CE. When refined with full set of atoms, > everything but CE ends up pretty much in the same place. I am sure > terminal torsion angle affects positioning of atoms that are closer > to the backbone, but the differences might be below the margin of > error and there is simply no evidence for multiple positions of > these. > > If you check, for instance, standard conformers 2, 3 and 4 for > methinonine in coot, they are pretty much the same except for CE. > It seems reasonable that only CE should be modeled in alternative > positions. > > Cheers, > > Ed > > On Wed, 2012-06-06 at 23:23 +0200, Tim Gruene wrote: Dear Ed, > > what resolution do your data have which make you think that the > map you are staring at is without model bias? It is worth splitting > the whole side chain for dual conformations and unlikely that > really only one single atom as two possible conformations. > > Regards, Tim > > On 06/06/2012 09:16 PM, Ed Pozharski wrote: >>>> Are there plans to implement a "partial" double conformer >>>> feature? What I mean by this is that currently one can only >>>> choose whether to make alternate conformers with or without >>>> backbone atoms. I am staring at a methionine which only >>>> appears to have CE in different position, so it does not seem >>>> reasonable to have alternate positions for every other side >>>> chain atom. >>>> >>>> The workaround is to create double conformer for the whole >>>> side chain and then delete individual atoms - 4 clicks in >>>> this case, quite doable. But I thought maybe there is some >>>> trimming option or ideally one would be asked at which atom >>>> to split the chain. >>>> >>>> Cheers, >>>> >>>> Ed. >>>> > > - -- Dr Tim Gruene Institut fuer anorganische Chemie Tammannstr. 4 D-37077 Goettingen GPG Key ID = A46BEE1A -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iD8DBQFP0FWsUxlJ7aRr7hoRAkGxAJ9ka6+5CBwzwzVPm0Vg44M8oJDnhQCg8Lv5 MruGeZt+KeOCjgP7Ei0yzIY= =56G4 -----END PGP SIGNATURE-----
