[phenixbb] How to eliminate or decrease negative electron density at ligand position
pafonine at lbl.gov
Wed Oct 13 16:38:06 PDT 2010
> I have fine electron density 2Fo-Fc around a small ligand (~20 atoms)
> binding to my protein (at 1sigma level, all of the ligand atoms are
> surround by electron density, resolution=3 angstrom). However, I also
> see some negative electron density Fo-Fc surrounding some of the
> ligand atoms. My question is should I care about these negative
> electron density? If I should, what should I do in order to decrease
> or eliminate these negative electron density? I have tried occupancy
> refinement. It does help to decrease the negative density. However, it
> also makes 2Fo-Fc weaker. Another method I tried is to use tls
> refinement only for the ligand. It also helped to decrease the
> negative electron density. But since the data resolution is only 3
> angstrom, I do not know if I am even allowed to do tls refinement?
> Thank you all in advance.
Yes, Jason, you should care about it, and that should prompt you to
think about possible refinement options for your ligand.
1) Your ligand is probably partially occupied, so you may need to refine
its overall occupancy;
2) the ligand my be in alternative locations, so you may need to account
for this in refinement;
3) Since the data resolution is 3A, the above "1" and "2" are tricky to
do fairly, so you may and should try but read the results skeptically.
If you send me the data and model I will be happy to play with it and
suggest any help I can. The only caveat is that I'm traveling right now
and will be able to do this only after October 20. So, it's up to you -
if you can wait then just send me the files and I will do my best to
help you (after October 20).
Otherwise we can keep chatting about this to figure out something that
you can do meanwhile and that may fix his problem.
Please let me know what you prefer, I'm ok with either choice -:) Well,
at leat you can do "1)" and "2)" right now while I'm sleeping in
All the best!
More information about the phenixbb