I guess I have known lots of incorrectedly fitted ligands, and am never sure quite how to be confident they are correct. Your resolution is good, so one hopes the density is pretty clear? Common sources of error. By far the most common is caused by having an incorrect dictionary. Look very carefully at what is expected - is the chirality correct; the planarity? the bonding - does it fit what the chemists expect??
Second - over interpreting the density - some ligands have extremely wobbly bits, and I am never sure what to do about them - I tend to get overoptimistic, but with your data the B factors should help indicate such sections. I set the occupancies to 0.00 for those bits - re-refine and look at the difference maps again for verification - coot is very good to fitting ligands if there is density to fit to. Make sure the B factor restraints within the ligand are not too tight . Sometimes if your ligand contains S or P and the data is good enough you can use the anomalous difference fourier maps to see if there are peaks over the expected scatterers. (This check is a bit of a pain to do - you have to use CAD to add DANO SIGDANO to the REFMAC output, then do the map and peak search) Eleanor On 12 March 2013 10:02, <herman.schreu...@sanofi.com> wrote: > ** > Dear Srinivasan, > > The first thing I would do is to look very carefully at the electron > density maps: Does it look like a bunch of water molecules, or is > continuous density present? Could it be some buffer component (Tris, Hepes, > sulfate, DMSO etc) or the counter ion of your ligand or the cryoprotectant > (e.g. glycerol)? Does the protein have disordered residues e.g. at the N- > or C-terminus, which could bind to the binding site? Same is true for > sugars if your protein is glycosylated. If you are crystallizing a > protease, autolysis during crystallization may have released short peptides > which may bind to the ligand binding site. Is the binding mode of your > ligand disordered? Would the electron density be better explained if you > fit the ligand in two alternative conformations? If your ligand is bound at > low occupancy, something else (e.g. waters, sulfate) may be bound in those > binding sites where the ligand is not present. Also side chains may adopt a > different conformation in the absence of the ligand. In this case one > should model all alternatives. > > After this analysis, one usually ends up with a limited number > of alternatives (e.g. ligand is bound, bunch of water molecules, something > else is bound). Most informative is to fit and refine all (in this case > three) possibilities and look which electron density map looks most > convincing. I would also calculate the real space correlation coefficient > for each alternative. If possible, I would also compare the electron > density with the electron density of apo-crystals (not cocrystallized or > soaked with the ligand). If the same density is present there, it is not > your ligand but something else. > > Good luck! > Herman > > ------------------------------ > *From:* CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] *On Behalf Of * > R.Srinivasan > *Sent:* Monday, March 11, 2013 11:03 PM > > *To:* CCP4BB@JISCMAIL.AC.UK > *Subject:* [ccp4bb] validating ligand density > > Hello all, > > We co-crystallized an inactive variant of our enzyme in the > presence of substrate and have determined the structure at 1.85A. > > Now, we want to validate the fitting of the ligand into the > electron density. We tried validating using the difference map (2Fo-Fc) > after refining the structure without the ligand. But, it is still a bit > inconclusive if the density fits the ligand. > > It would be very kind to know if there are tools for > validating this electron density. We were excited about twilight but turns > out it can only be used with deposited structure. > > We will appreciate your help and suggestions. > > Many thanks, > Srinivasan > >
