Dear Mariah,
Recently I significantly improved an alpha-helical structure using
"restraints" in Phenix. Maybe, this will help you to improve your model
and get better Pymol "pictures". I copy the message I got from Phenix
people here, as many CCP4 users may also face such a situation.
Hi Stefan,
> > Is it possible to restrain refinement according to secondary >
> structures in phenix?
Yes, although it is a bit tedious at the moment.
See the "Definition of custom bonds and angles" section here:
https://www.phenix-online.org/download/documentation/phenix/development/refinement.htm
You will need some way to figure out the hydrogen bonds (e.g. use DSSP,
http://swift.cmbi.ru.nl/gv/dssp/). Then re-format the information
like this:
refinement.geometry_restraints.edits {
bond {
action = *add
atom_selection_1 = resname XXX and resid 123 and name X
atom_selection_2 = resname YYY and resid 124 and name Y
distance_ideal = 2.1
sigma = 0.02
slack = 0.2
}
bond {
action = *add
atom_selection_1 = resname XXX and resid 125 and name X
atom_selection_2 = resname YYY and resid 126 and name Y
distance_ideal = 2.1
sigma = 0.02
slack = 0.2
}
}
Ralf
Regards,
Stefan
Charlie Bond a écrit :
Dear Maria,
I can see that others have suggested how to manually change the
secondary structure, but I have one comment:
If a secondary structure detection algorithm doesn't assign those
residues as 'strand' then this is probably (bugs notwithstanding) a
feature of your model and you should also take care to look that the
local structure (hydrogen bonding etc) makes sense at that point.
While we shouldn't blindly accept the output of computer programs,
altering their output to fit our preconceptions isn't necessarily a
good idea either.
Cheers,
Charlie
protein.chemist protein.chemist wrote:
Dear All,
I have a question about PYMOL.
I am trying to make a cartoon diagram of my protein showing the
secondary structural elements. However for some reasons it is not
showing the N terminal strand and showing that as a loop. Is there
any command for selecting a range of residues and showing them as a
strand or helix.
Thanks for helping.
Mariah
--
Mariah Jones
Department of Biochemistry
University of Florida
