Re: [ccp4bb] brute force molecular replacement
Dear Xie, > What are the other brute force programs for molecular replacement out > there? Qs (available via http://www.mbg.duth.gr/~glykos/Qs.html) can be as brutal with your CPU(s) as they can take. Nicholas -- Dr Nicholas M. Glykos, Department of Molecular Biology and Genetics, Democritus University of Thrace, University Campus, 68100 Alexandroupolis, Greece, Fax +302551030613 Tel ++302551030620 (77620), http://www.mbg.duth.gr/~glykos/
[ccp4bb] more on lattice translocation defects
May I inquire if someone has a copy of Zhu et. al. (Acta Cryst. D. 2008 D64, 843-850) that might be posted somewhere free? I don't have subscription to Acta Cryst D and this one is possibly a problem I have in a structure and I don't understand what it means exactly. Or if someone can give a definition in terms that is simple (for my simple mind). Thanks -- Laurie Betts X-ray Crystallography Facility Manager Department of Structural Biology University of Pittsburgh 1050 BST3 3501 Fifth Ave Pittsburgh, PA 15260 412-383-5839
[ccp4bb] Problem during refinement
Dear All, I am refining a structure (resolution 2.4A) using Refmac and am getting a lot of cis-peptides (about 16 for prolines as well as other amino acids). There are some side chain swaps as well leading to high differences Rfactor: 20.6 and Rfree: 27.6. Although I tried to convert them to trans form and then tried to refine but it shows them again as cis peptides only. In order to rectify the second problem of side chain swaps I have tried to mutate back the residues followed by refinements but it is also of no use. I need to know if anyone of you have ever encountered these kind of problems. I will be greatly helped by your suggestions in this regard. Best regards, Anil Kumar (Graduate Student) - School of Biological Sciences Division of Structural & Computational Biology Nanyang Technological University, 60 Nanyang Drive Singapore 639798. anil0...@ntu.edu.sg Tel: +65-6316-2926 - Check out the all-new Messenger 9.0! Click here.
Re: [ccp4bb] Problem during refinement
IT could be miscommunication between refmac and coot. Try to remove cispep lines from your pdb and refine again. Then it may work. After model building info about cispep and/or ssbridge should be removed from the pdb file (unless you are confident about them) regards Garib On 25 Jan 2009, at 19:32, anil kumar wrote: Dear All, I am refining a structure (resolution 2.4A) using Refmac and am getting a lot of cis-peptides (about 16 for prolines as well as other amino acids). There are some side chain swaps as well leading to high differences Rfactor: 20.6 and Rfree: 27.6. Although I tried to convert them to trans form and then tried to refine but it shows them again as cis peptides only. In order to rectify the second problem of side chain swaps I have tried to mutate back the residues followed by refinements but it is also of no use. I need to know if anyone of you have ever encountered these kind of problems. I will be greatly helped by your suggestions in this regard. Best regards, Anil Kumar (Graduate Student) - School of Biological Sciences Division of Structural & Computational Biology Nanyang Technological University, 60 Nanyang Drive Singapore 639798. anil0...@ntu.edu.sg Tel: +65-6316-2926 Check out the all-new Messenger 9.0! Click here.
Re: [ccp4bb] cbuccanner.exam failed with libFFTW generated by Intel MKL
Dear Kevin, Thanks for your advice. I compared cpirate.exam log between pre-built and intel-FFT linked. the log values seems distinctly different. Maps with pre-build MLPHARE and intel-FFT don't seem so different. So, I tried the comparison using phase which estimated by pre-build mlphare.exam. However, cpirate.exam's logs are different. As a result, pre-build cbuccarneer.exam is successful and intel-FFT is failed. So intel-FFT library may act differently in this case. For reference, the logs are attached. --- the cpirate.exam logs by pre-built vs intel-FFT Automatic content fitting: Skew by: -0.21875,0 (dense, ordered) 0.96875,0.875 (dense, ordered) Unbiased results from initial cycle: R-factor : 0.790483 0.195488 Free R-factor: 0.845515 1.15188 E-correl : 0.504196 0.991128 Free E-correl: 0.195072 0.2489 LgLkGain : 0.331314 2.8692 Free LgLkGain: 0.30185-inf Biased results from cycle 2: Biased R-factor : 0.6762870.140585 Biased Free R-factor: 0 0 Biased E-correl : 0.6507680.988193 Biased Free E-correl: 0 0 Biased LgLkGain : 0.526848nan Biased Free LgLkGain: 0 0 Biased results from cycle 3: Biased R-factor : 0.60964 0.191351 Biased Free R-factor: 0 0 Biased E-correl : 0.7269250.977604 Biased Free E-correl: 0 0 Biased LgLkGain : 0.694419nan Biased Free LgLkGain: 0 0 Thanks, Nobuo OKAZAKI Kevin Cowtan wrote: > OK, I had a closer look at this. > > Buccaneer is failing because it can't grow a single residue of chain > from the seed positions. This can only happen if the map is total garbage. > > So the first thing to do is check the phases which are going into > buccaneer in both the gcc and intel cases. Calculate a map from each > set, and make sure they are both reasonably similar and interpretable > (they should be for the example data). > > If the two maps, differ, then the problem is in the upstream program. > > If they don't, then the most likely cause is that buccaneer is not > calculating the map correctly. Add 'mapout buc.map' to the buccaneer > command file, and rerun both the gcc and intel versions. If the > buccaneer map from the intel version is corrupted, then the problem is > most likely the linking of the fft library in the intel case.
Re: [ccp4bb] cbuccanner.exam failed with libFFTW generated by Intel MKL
I have checked CLIPPER with lib/clipper/examples/test.csh and compared logs between gcc and intel-FFT. "diff" output about 1000 different lines. These are few lines of output. (intel-FFT) < HKL = ( 2, 2, 0) 198 306:8.25 306: 198 306 < HKL = ( 2, 2, 1) 199 153:8.29 153: 199 153 < HKL = ( 2, 2, 2) 200 360:8.33 360: 200 360 (gcc) > HKL = ( 2, 2, 0) 198 306: 198 306: 198 306 > HKL = ( 2, 2, 1) 199 153: 199 153: 199 153 > HKL = ( 2, 2, 2) 200 360: 200 360: 200 360 It suggested CLIPPER with Intel-FFT is not all right. Thanks, Nobuo OKAZAKI
