Thanks everyone for your valuable suggestions
On 4 February 2014 20:47, jai mohan <[email protected]> wrote: > Dear Shanti Pal Gangwar, > You are trying to solve your crystal structure of what! whether its a > peptide or protein or enzyme or nucleic acid? There are several reasons > behind getting a high Rmerge, it not only depends on the data quality. What > are all the efforts! that you have applied to solved your crystal > structure. Provide some more information like, > > -atleast some preliminary info of data collection methods > -the program(s) you tried to process your data set > -space group/unit cell length/completeness for the low and high resolution > shells/ I\s. > -whether your molecule is a monomer/dimer/......in asymmetric unit > -have you access the data quality (very easy in Xtriage-PHENIX), whether > it contains any twinning / translational pseudosymmetry. > -some info about your molecular replacement programs(s) output like > correlation coefficient, Rfactor | LLG, RF, TF, PAK | > -the refinement program(s) you tried > -hows your electron density map look like after the initial rounds of > refinement > > you could easily include all the above information within a two or maximum > three lines. If your question is very clear than definitely you will get > more inputs from the experts, so that you will be guided in a right way to > get your crystal structure ASAP! > > Regards, > Dr. S. M. Jaimohan > > > On Tuesday, 4 February 2014 8:18 PM, Bernhard Rupp < > [email protected]> wrote: > ...the slash at the end is obviously the secret sign for > "google-translate into a randomly picked language" > > http://scripts.iucr.org/cgi-bin/citedin?dz5235 > > > *From:* Rick Lewis [mailto:[email protected]] > *Sent:* Dienstag, 4. Februar 2014 15:24 > *To:* [email protected] > *Subject:* Re: [ccp4bb] high Rwork / Rfree after MR > > a third of nothing is till nothing... see attached screendump > > ;-) > On 04/02/14 14:19, Bernhard Rupp wrote: > > EDSTATS is part of CCP4 > > ....which in fact I cite in 33% of all of its CrossRef references > > http://scripts.iucr.org/cgi-bin/citedin?dz5235/ > > Touché. > > *From:* CCP4 bulletin board > [mailto:[email protected]<[email protected]>] > *On Behalf Of *Frank von Delft > *Sent:* Dienstag, 4. Februar 2014 14:40 > *To:* [email protected] > *Subject:* Re: [ccp4bb] high Rwork / Rfree after MR > > And really we should not be using real space CC or real space R anymore - > not since Tickle 2012 ( > http://scripts.iucr.org/cgi-bin/paper?S0907444911035918) > > EDSTATS is part of CCP4. > > phx > > > > > > On 04/02/2014 13:31, Bernhard Rupp wrote: > > > interpretable electron density map is obtained to which most of your > protein model can be fitted properly. The primary numeric indication for > that are your R-factors. > > Minor comment: The measure for the fit of the **model (density) to the > electron density** is actually a local real space correlation coefficient > or a local real space R value, not the global reciprocal space R-values > (not a **factor** in the mathematical sense, nor a factor for anything > else). The overall linear residual between model structure factor > amplitudes Fc and observed amplitudes Fo (aka R(F)-value) is a global > reciprocal space measure, which can be affected by many other contributions > causing the misfit between Fo and Fc data, in addition to bad model, as > already noted below. > > Ø The high values you have suggest that either something is wrong with > your space group, twinning, etc. as Eleanor suggested, or your space group > and MR solution may be still correct but for a search model of poor > similarity with your crystal structure. Even in the later case, at this > resolution it is going to be very challenging to build and refine a > structure to reasonable R-factors starting from a poor model. > > Hay > > > ======================================== > Hay Dvir Ph. D. > Head Technion Center for Structural Biology > Technion Haifa 3200003, Israel > Tel: +(972)-77-887-1901 > Fax: +(972)-77-887-1935 > E-mail [email protected] > Website http://tcsb.technion.ac.il/Hay-Dvir > ======================================== > > > > On Feb 4, 2014, at 11:36 AM, Shanti Pal Gangwar wrote: > > > > > Dear All > I have solved one structure by MR. The data data quality was poor so the > Rmerge was high. The resolution of the data is 3.3 Angstrom.The refinement > statistics are also very poor Rw/Rf= 40/42 %. After many efforts we are not > able to get reasonable Rw/Rf. > > My question is "can it be claimed that we have solved the structure?" > > > > Thanks in advance > > > > > > -- > ******************** > regards > Shanti Pal Gangwar > School of Life Sciences > Jawaharlal Nehru University > New Delhi-110067 > India > Email:[email protected] > > > > > > > > > > -- > > R. J. Lewis > > Professor of Structural Biology > > Institute for Cell and Molecular Biosciences > > Faculty of Medical Sciences Tel: +44 (0)191 222 5482 > > University of Newcastle Fax: +44 (0)191 222 7424 > > Newcastle upon Tyne, NE2 4HH, UK Email: [email protected] > > URL: sbl.ncl.ac.uk > > > > > > -- ******************** regards Shanti Pal Gangwar School of Life Sciences Jawaharlal Nehru University New Delhi-110067 India Email:[email protected]
