Dear All, What I read from the email is that the R/Rfree values do not change much if you refine with high occupancy-high B or with low occupancy-low B, which is what one expects with 2.6 Å data. This is something different from leaving out the inhibitor from the refinement.
Considering the affinity, one can not compare the affinity free in solution with the affinity in a crystal. Crystal packing may hinder peptide binding. Another, very important thing is to add sufficient peptide to the cryoprotectant, otherwise one may loose almost all bound peptide during the cryoprotection. What often works for us, is just trying another crystal, we sometimes find very different inhibitor-occupancies for (almost) identical conditions. Best regards, Herman -----Ursprüngliche Nachricht----- Von: CCP4 bulletin board [mailto:[EMAIL PROTECTED] Im Auftrag von Manfred S. Weiss Gesendet: Montag, 30. April 2007 16:32 An: CCP4BB@JISCMAIL.AC.UK Betreff: Re: [ccp4bb] extra high B factor Dear all, Frankly speaking I am having some doubts about this whole discussion. 1. Apparently, it does not make a difference in R and Rfree whether the peptide is in the structure or not. This does suggest that there is very little (if any) information about the peptide in the data. Right? 2. You stated that you can unambiguously trace the peptide from N- to C-terminus. If you assume atomic B-factors of 100 or larger and calculate the density, I really can't see how this is possible. Maybe you could produce an omit S. A. difference electron density map to show this. 3. You said the affinity between protein and peptide is 10^-7 or 10^-8 M. This means that something is a bit strange in any case. With this affinity you should get 100% occupancy. Of course, it is possible that your buffer/cryo-solvent/etc. reduce the affinity. Have you considered this? Cheers, Manfred. ******************************************************************** * * * Dr. Manfred S. Weiss * * * * Team Leader * * * * EMBL Hamburg Outstation Fon: +49-40-89902-170 * * c/o DESY, Notkestr. 85 Fax: +49-40-89902-149 * * D-22603 Hamburg Email: [EMAIL PROTECTED] * * GERMANY Web: www.embl-hamburg.de/~msweiss/ * * * ******************************************************************** On Mon, 30 Apr 2007, Jiamu Du wrote: > Does anyone know a program can perform the ocupancy refinement? > Or we always only refine B factor to reflect the occupancy? > > Thanks > > > On 4/30/07, Eleanor Dodson <[EMAIL PROTECTED]> wrote: > > > > Well - it is extremely likely that the peptide is partially occupied > > and the occupancy may well be < 0.5.. > > > > But at this resolution you are going to have great difficulty > > deciding whether you should have Occ=1.0 <B = 130> > > > > Occ = 0.5 <B = 100> > > > > Occ = 0.33 <B = ??? 80???> > > > > As your Rfactors show it makes very little difference to any scoring > > system.. > > > > You can look at difference maps and try to see if one looks flatter > > than the other .. > > > > Even the overall Wilson plot B is not very well determined, so I > > wouldnt worry too much.. > > > > Eleanor > > > > Jiamu Du wrote: > > > Dear All: > > > According to your suggestion, I have set the peptide's occupency > > > to 0.5. Two strategies were employed. > > > 1. Direct using Refmac restrained refinement for 10 cycles. The B > > > factor only drops to around 100. R/Rf did not change, either. > > > 2. Direct CNS B-fator refinemen. The B factor drops to a moderate > > > level 60-80, and the R/Rf each increases about 2%. > > > 3. First using CNS B-fator refinemen nad next Refmac restrained > > > refinement. The B factor drops to 60-80, and the R/Rf did not change. > > > > > > I think next step TLS refinement should be carried out. > > > > > > > > > On 4/30/07, *Philippe DUMAS* <[EMAIL PROTECTED] > > > <mailto:[EMAIL PROTECTED]>> wrote: > > > > > > Jiamu > > > > > > According to the numbers you have mentioned I conclude that you > > > peptide occupancy should be around 60-64 % > > > I am interested to know what will be the value that you will > > > obtain after refinement... > > > > > > > > > Philippe Dumas > > > IBMC-CNRS, UPR9002 > > > 15, rue René Descartes 67084 Strasbourg cedex > > > tel: +33 (0)3 88 41 70 02 > > > [EMAIL PROTECTED] <mailto:[EMAIL PROTECTED]> > > > > > > > > > -----Message d'origine----- > > > *De :* CCP4 bulletin board [mailto: CCP4BB@JISCMAIL.AC.UK > > > <mailto:CCP4BB@JISCMAIL.AC.UK>]*De la part de* Jiamu Du > > > *Envoyé :* lundi 30 avril 2007 05:57 > > > *À :* CCP4BB@JISCMAIL.AC.UK <mailto:CCP4BB@JISCMAIL.AC.UK> > > > *Objet :* [ccp4bb] extra high B factor > > > > > > Dear All: > > > I am refining a protein-peptide complex struture at 2.6 > > > angstrom resolution. > > > The data was obtain from a co-crystal and the wilson B factor > > > of the data is about 70. > > > The affinity between protein and peptide is about 10E-7 to > > > 10E-8 molar. > > > Protein fragment of the structure has a common B facor about 50. > > > But surprisingly, the average B factor of the peptide is as > > > high as 130, although the peptide can be clearly traced from > > > the the electron density map. All residues of the peptide have > > > such a high B factor. > > > My question is how can I reduce the abnormal high B factor to > > > a common level or if this high B factor acceptable. > > > And another question is if this high B fator will influence > > > the final refiment level. > > > > > > Thanks. > > > > > > -- > > > Jiamu Du > > > State Key Laboratory of Molecular Biology > > > Institute of Biochemistry and Cell Biology Shanghai Institutes > > > for Biological Sciences > > > Chinese Academy of Sciences (CAS) > > > > > > > > > > > > > > > -- > > > Jiamu Du > > > State Key Laboratory of Molecular Biology Institute of > > > Biochemistry and Cell Biology Shanghai Institutes for Biological > > > Sciences Chinese Academy of Sciences (CAS) > > > > > > > -- > Jiamu Du > State Key Laboratory of Molecular Biology > Institute of Biochemistry and Cell Biology Shanghai Institutes for > Biological Sciences > Chinese Academy of Sciences (CAS) >
