Re: [ccp4bb] iMosflm can't read images
Dear Huiying, my first guess is that you copied the files onto an external hard drive which was formatted for Windows (with vfat). If you tried to read the images from the same drive can confuse Linux systems, because Linux is literate, whereas vfat does not distinguish between capital and lower case letters. If that's a possible reason try copying the images to a local drive and check the permissions (which also often get screwed up by copying files to and fro vfat). Tim On Wed, Sep 22, 2010 at 03:23:37PM -0700, Huiying Li wrote: I had trouble adding image files into iMosflm (1.0.4) GUI running on our in-house Linux machine (64-bit, Fedoro). It complained about the filenames: * FATAL ERROR * Image filenames must be of the form ABCDE_###.ext or ABCDE-###.ext where the initial string can be up to 40 characters long and must be separated from a 3 digit number by a _ or -, and the extension (ext) can be up to 8 characters long. The filenames are actually following the convention Mosflm required. The same image files were readable by iMosflm on Linux machine at the synchrotron station. Is anything wrong with our iMosflm installation, configuration...? Thanks for your help. Huiying _ Huiying Li, Ph. D Department of Molecular Biology and Biochemistry University of California at Irvine Irvine, CA 92697, USA Tel: 949-824-4322(or -1953); Fax: 949-824-3280 email: h...@uci.edu -- -- Tim Gruene Institut fuer anorganische Chemie Tammannstr. 4 D-37077 Goettingen GPG Key ID = A46BEE1A signature.asc Description: Digital signature
Re: [ccp4bb] Effect of NCS on estimate of data:parameter ratio
Hi Gerard Pavel Isn't this the proviso I was referring to, that one cannot in practice use an infinite weight because of rounding errors in the target function. The weight just has to be 'big enough' such that the restraint residual becomes sufficiently small that it's no longer significant. In numerical constrained optimisation the method of increasing the constraint weights (a.k.a. 'penalty coefficients') until the constraint violations are sufficiently small is called the 'penalty method', see http://en.wikipedia.org/wiki/Penalty_method . The method where you substitute some of the parameters using the constraint equations is called (you guessed it!) the 'substitution method', see http://people.ucsc.edu/~rgil/Optimization.pdf . There are several other methods, e.g. the 'augmented Lagrangian method' is very popular, see http://www.ualberta.ca/CNS/RESEARCH/NAG/FastfloDoc/Tutorial/html/node112.html . As in the penalty method, the AL method adds additional parameters to be determined (the Lagrange multipliers, one per constraint) instead of eliminating some parameters using the constraint equations; however the advantage is that it removes the requirement that the penalty coefficient be very big. The point about all these methods of constrained optimisation is that they are in principle only different ways of achieving the same result, at least that's what the textbooks say! And now after the penalties and substitutions it's time to blow the whistle ... Cheers -- Ian On Wed, Sep 22, 2010 at 10:00 PM, Pavel Afonine pafon...@lbl.gov wrote: I agree with Gerard. Example: it's unlikely to achieve a result of rigid-body refinement (when you refine six rotation/translation parameters) by replacing it with refining individual coordinates using infinitely large weights for restraints. Pavel. On 9/22/10 1:46 PM, Gerard DVD Kleywegt wrote: Hi Ian, First, constraints are just a special case of restraints in the limit of infinite weights, in fact one way of getting constraints is simply to use restraints with very large weights (though not too large that you get rounding problems). These 'pseudo-constraints' will be indistinguishable in effect from the 'real thing'. So why treat restraints and constraints differently as far as the statistics are concerned: the difference is purely one of implementation. In practice this is not true, of course. If you impose infinitely strong NCS restraints, any change to a thusly restrained parameter by the refinement program will make the target function infinite, so effectively your model will never change. This is very different from the behaviour under NCS constraints and the resulting models in these two cases will in fact be very easily distinguishable. --Gerard ** Gerard J. Kleywegt Dept. of Cell Molecular Biology University of Uppsala Biomedical Centre Box 596 SE-751 24 Uppsala SWEDEN http://xray.bmc.uu.se/gerard/ mailto:ger...@xray.bmc.uu.se ** The opinions in this message are fictional. Any similarity to actual opinions, living or dead, is purely coincidental. **
[ccp4bb] Postdoctoral position in membrane protein crystallography
We are searching for outstanding postdoctoral candidates that wish to pursue structure-function studies on membrane proteins at the Vanderbilt University Medical Center. We are focused on the determination of structures of membrane proteins in complex with soluble binding partners, and are also interested in furthering the development of techniques for membrane protein crystallization. Prior experience in protein expression and purification would aid in pursuing these projects, as would experience in x-ray crystallography. The Center for Structural Biology at the Vanderbilt University Medical Center has excellent resources for pursuing structural studies. The on-campus x-ray facilities include an R-AXIS IV system, a new high-brilliance Bruker x-ray generator with a CCD detector, and an xcalibur system. All of these sources are equipped with cryogenic capability for data collection. A state-of-the art robotic crystallization facility includes a Mosquito nano-liter pipetting robot, a Hamilton Star-let liquid handler, a cubic lipid phase pipetting robot, and a Formulatrix imager. Vanderbilt University has purchased a share in a beamline at the Advanced Photon Source which provides us with an average of 24 hours on 3 beamlines concurrently every 3 weeks. Vanderbilt University is located in Nashville, an exciting multi-cultural center that houses the music industry in this region. The mild climate makes a pleasant background for this diverse and lively city. Nature lovers will enjoy the nearby Smoky Mountains and find other outdoor recreation easily accessible. Qualified applicants should send a cover letter and cv by email to: tina.iver...@vanderbilt.edu. - T. M. Iverson, Ph.D. Department of Pharmacology 460 Robinson Research Building 23rd Avenue South @ Pierce Vanderbilt University Medical Center Nashville, TN 37232-6600 Office: (615) 322-7817 Fax: (615) 343-6532 Email: tina.iver...@vanderbilt.edu Lab URL: http://www.mc.vanderbilt.edu/root/vumc.php?site=iversonlab Department URL: http://www.vanderbilt.edu/pharmacology
[ccp4bb] Protein melting temperatures
Hi Folks, Sorry for the pre-xtallo question; pre-xtallo right now, but hoping to take my protein the xtallo way one of these days! I am currently performing Thermofluor assays with my protein and the results show that the Tm is ~45C. I am looking for some examples of proteins and their melting temperatures so that I can gauge where my protein falls in the spectrum of unstable-to-stably folded. For example, the melting temperature of some forms of lysozyme is 73.8C (very stable, I suppose). Just need a sense for whether my protein is considered unstable or somewhat stable. Please could you share some examples. Many thanks. Raji --- Raji Edayathumangalam Joint Research Fellow Harvard Medical School/ Brigham and Women's Hospital Brandeis University
Re: [ccp4bb] Protein melting temperatures
There is a nice paper Comput Biol Chem. 2009 Dec;33(6):445-50. Epub 2009 Oct 20. Predicting melting temperature directly from protein sequences. Ku T, Lu P, Chan C, Wang T, Lai S, Lyu P, Hsiao N. They have a list of 35 different proteins with their Tms with the references from where they obtained their data. Hope this aids in your work. Dan
[ccp4bb] Positions Available in Membrane Protein Structural Biology
Postdoctoral Positions: Membrane Proteins @ University of Virginia Due to some recent changes in both funding and personnel, I have several positions available immediately in my lab. Postdoctoral level preferred, but more senior (or more junior) positions are possible. Relevant experience is certainly desirable. Current areas of research include bacterial outer membrane active transport, eukaryotic/mammalian membrane protein structural genomics, and methods development (for membrane protein structural biology). The scientific and intellectual environment for membrane research is very good. Eight labs working in membrane protein biochemistry, biophysics and structural biology are sited in a modern research building. The Center for Membrane Biology includes additional investigators, holds a regular seminar series, and maintains a cohesive, lively and collegial environment. Applicants, please send me a cover letter, cv, and names of three references. Also, any other queries can be directed to me, as well. Michael Wiener, 434-243-2731, mwie...@virginia.edu [The University of Virginia is an Equal Opportunity/Affirmative Action Employer.]
Re: [ccp4bb] Effect of NCS on estimate of data:parameter ratio
Dear All one thing I remembered from what Gerard pointed out was the difference in the XPLOR/CNS formalism between strict and restrained which is not a continuum. Restrained was obviously when you had multiple copies and they were restrained with a weight (which was like a force constant) to be similar when superimposed. So if you increase the force constant then they can move during refinement but they all try to move together when they move. And the other extreme is strict where there was no force applied at all but only a single copy of the chain and the ASU is built by applying the NCS symmetry. The atoms are free to move but, unlike the case with restrained where there is superimposition on the fly, in the strict case there is no automatic update of the superimposition matrices. So every move gets religiously copied to all the chains when the ASU is made. At this point I guess the copies can bump and so apply a force on each other but that is a local, and likely to be perturbing, force. best wishes Martyn Martyn Symmons Cambridge --- On Thu, 23/9/10, Ian Tickle ianj...@gmail.com wrote: From: Ian Tickle ianj...@gmail.com Subject: Re: [ccp4bb] Effect of NCS on estimate of data:parameter ratio To: CCP4BB@JISCMAIL.AC.UK Date: Thursday, 23 September, 2010, 11:21 Hi Gerard Pavel Isn't this the proviso I was referring to, that one cannot in practice use an infinite weight because of rounding errors in the target function. The weight just has to be 'big enough' such that the restraint residual becomes sufficiently small that it's no longer significant. In numerical constrained optimisation the method of increasing the constraint weights (a.k.a. 'penalty coefficients') until the constraint violations are sufficiently small is called the 'penalty method', see http://en.wikipedia.org/wiki/Penalty_method . The method where you substitute some of the parameters using the constraint equations is called (you guessed it!) the 'substitution method', see http://people.ucsc.edu/~rgil/Optimization.pdf . There are several other methods, e.g. the 'augmented Lagrangian method' is very popular, see http://www.ualberta.ca/CNS/RESEARCH/NAG/FastfloDoc/Tutorial/html/node112.html . As in the penalty method, the AL method adds additional parameters to be determined (the Lagrange multipliers, one per constraint) instead of eliminating some parameters using the constraint equations; however the advantage is that it removes the requirement that the penalty coefficient be very big. The point about all these methods of constrained optimisation is that they are in principle only different ways of achieving the same result, at least that's what the textbooks say! And now after the penalties and substitutions it's time to blow the whistle ... Cheers -- Ian On Wed, Sep 22, 2010 at 10:00 PM, Pavel Afonine pafon...@lbl.gov wrote: I agree with Gerard. Example: it's unlikely to achieve a result of rigid-body refinement (when you refine six rotation/translation parameters) by replacing it with refining individual coordinates using infinitely large weights for restraints. Pavel. On 9/22/10 1:46 PM, Gerard DVD Kleywegt wrote: Hi Ian, First, constraints are just a special case of restraints in the limit of infinite weights, in fact one way of getting constraints is simply to use restraints with very large weights (though not too large that you get rounding problems). These 'pseudo-constraints' will be indistinguishable in effect from the 'real thing'. So why treat restraints and constraints differently as far as the statistics are concerned: the difference is purely one of implementation. In practice this is not true, of course. If you impose infinitely strong NCS restraints, any change to a thusly restrained parameter by the refinement program will make the target function infinite, so effectively your model will never change. This is very different from the behaviour under NCS constraints and the resulting models in these two cases will in fact be very easily distinguishable. --Gerard ** Gerard J. Kleywegt Dept. of Cell Molecular Biology University of Uppsala Biomedical Centre Box 596 SE-751 24 Uppsala SWEDEN http://xray.bmc.uu.se/gerard/ mailto:ger...@xray.bmc.uu.se ** The opinions in this message are fictional. Any similarity to actual opinions, living or dead, is purely coincidental. **
Re: [ccp4bb] Protein melting temperatures
Not sure whether this is the kind of information you are looking for: The protein with PDB-ID 1ofc had a melting temperature of 37deg (from CD), which was supported by the fact that it did not express in E.coli at that temperature. At 20deg it expressed to about 60mg / (liter LB), could be concentrated to more than 100mg/ml, crystallised at room temperature and diffracted to 1.9A. The initial purification steps were done at 4deg, but I guess that's generally good advice anyhow. So maybe you don't need to worry too much, because stability is probably not the same as thermal stability. Cheers, Tim On Thu, Sep 23, 2010 at 11:28:20AM -0400, Raji Edayathumangalam wrote: Hi Folks, Sorry for the pre-xtallo question; pre-xtallo right now, but hoping to take my protein the xtallo way one of these days! I am currently performing Thermofluor assays with my protein and the results show that the Tm is ~45C. I am looking for some examples of proteins and their melting temperatures so that I can gauge where my protein falls in the spectrum of unstable-to-stably folded. For example, the melting temperature of some forms of lysozyme is 73.8C (very stable, I suppose). Just need a sense for whether my protein is considered unstable or somewhat stable. Please could you share some examples. Many thanks. Raji --- Raji Edayathumangalam Joint Research Fellow Harvard Medical School/ Brigham and Women's Hospital Brandeis University -- -- Tim Gruene Institut fuer anorganische Chemie Tammannstr. 4 D-37077 Goettingen GPG Key ID = A46BEE1A signature.asc Description: Digital signature
Re: [ccp4bb] Protein melting temperatures
Hello - The excellent paper of McCrary, uses differential scanning calorimetry, which will give an absolute measure of thermostability. Using Thermofluor I would be afraid you can only assess the relative thermostability of one protein in different conditions. As your fluorescence reporter would interact differently with exposed hydro[hobic patches in different proteins, I would be a bit more careful in comparing the Thermofluor results between different proteins ... I am not aware of anyone correlating differential scanning calorimetrywith Thermofluor data, but I must admit I have not looked up that literature recently. A. On 23 Sep 2010, at 18:40, Philippe DUMAS wrote: Le 23/09/2010 17:28, Raji Edayathumangalam a écrit : Raji I suggest having a look to this paper: McCrary et al. J. Mol. Biol. 264(1996) 784 where you will find an interesting study on protein stability and an interesting comparison with other proteins. Philippe Dumas Hi Folks, Sorry for the pre-xtallo question; pre-xtallo right now, but hoping to take my protein the xtallo way one of these days! I am currently performing Thermofluor assays with my protein and the results show that the Tm is ~45C. I am looking for some examples of proteins and their melting temperatures so that I can gauge where my protein falls in the spectrum of unstable-to-stably folded. For example, the melting temperature of some forms of lysozyme is 73.8C (very stable, I suppose). Just need a sense for whether my protein is considered unstable or somewhat stable. Please could you share some examples. Many thanks. Raji --- Raji Edayathumangalam Joint Research Fellow Harvard Medical School/ Brigham and Women's Hospital Brandeis University McCrary-JMB264(1996)784.pdfp_dumas.vcf
[ccp4bb] BMC errata updates
Dear All, first I want to thank all of you who kindly have sent me corrections, suggestions, and updates to BMC. Very helpful and much appreciated. In preparation for the e-book version and for the next printing, these corrections will now be incorporated by about End of October. If anyone has found more errors, small or big, or has other ideas for improvement, please let me know by that date. The current update and errata page is accessible at http://www.ruppweb.org/garland/errata.htm Best regards, BR - Bernhard Rupp 001 (925) 209-7429 +43 (676) 571-0536 b...@ruppweb.org hofkristall...@gmail.com http://www.ruppweb.org/ - No animals were hurt or killed during the production of this email. -
Re: [ccp4bb] Protein melting temperatures
I agree that you can't take two unrelated proteins and expect their Thermofluor Tms will be correlated with CD/DSC values. We've done quite a bit with point mutants, and it works well for that (see an example in our paper below). Also note that the dye is a perturbant the reduces the apparent Tm at higher concentrations, and of course that the whole point of using Thermofluor to help find Xtal conditions is that the apparent Tm is sensitive to buffer, ligands, etc. Tom http://www.chemistry.ohio-state.edu/~magliery/pdfs/LavinderMagliery2009JACS.pdf On 9/23/2010 1:03 PM, Anastassis Perrakis wrote: Hello - The excellent paper of McCrary, uses differential scanning calorimetry, which will give an absolute measure of thermostability. Using Thermofluor I would be afraid you can only assess the relative thermostability of one protein in different conditions. As your fluorescence reporter would interact differently with exposed hydro[hobic patches in different proteins, I would be a bit more careful in comparing the Thermofluor results between different proteins ... I am not aware of anyone correlating differential scanning calorimetrywith Thermofluor data, but I must admit I have not looked up that literature recently. A. On 23 Sep 2010, at 18:40, Philippe DUMAS wrote: Le 23/09/2010 17:28, Raji Edayathumangalam a écrit : Raji I suggest having a look to this paper: McCrary et al. J. Mol. Biol. 264(1996) 784 where you will find an interesting study on protein stability and an interesting comparison with other proteins. Philippe Dumas Hi Folks, Sorry for the pre-xtallo question; pre-xtallo right now, but hoping to take my protein the xtallo way one of these days! I am currently performing Thermofluor assays with my protein and the results show that the Tm is ~45C. I am looking for some examples of proteins and their melting temperatures so that I can gauge where my protein falls in the spectrum of unstable-to-stably folded. For example, the melting temperature of some forms of lysozyme is 73.8C (very stable, I suppose). Just need a sense for whether my protein is considered unstable or somewhat stable. Please could you share some examples. Many thanks. Raji --- Raji Edayathumangalam Joint Research Fellow Harvard Medical School/ Brigham and Women's Hospital Brandeis University McCrary-JMB264(1996)784.pdfp_dumas.vcf -- Thomas J. Magliery, Ph.D. Assistant Professor Department of Chemistry Department of Biochemistry The Ohio State University 1043 Evans Laboratory 100 West 18th Ave. Columbus, OH 43210-1185 (614) 859-5743 phone (Google Voice) (614) 292-1685 fax magli...@chemistry.ohio-state.edu http://www.chemistry.ohio-state.edu/~magliery
Re: [ccp4bb] Molecular replacement question
Maybe there is a domain shift of your protein compared to the model. If this is the case, try to do the MP with successive domains. 2010/9/13 Paul Holland pholl...@umd.edu Hello fellow crystallographers, I am trying molecular replacement for a protein crystal dataset that has very high sequence similarity to the search model with several predicted flexible loop regions; however, all attempts at finding a solution have not produce very ideal starting solutions using Phaser and Molrep (CC = 0.3 and Z-score = 5). I am very confident that the unit cell parameters are C2 84.027 120.565 108.272 90.00 104.71 90.00, and there appears to be no evidence of twinning. The Matthews calculation predicts from anywhere from 2-4 monomers in the ASU, and calculation of the SRF in Molrep does not identify any peaks in higher order symmetry except for the expected crystallographic two-fold for C2. Below is the table from the calculated SRF in molrep. Any advice would be greatly appreciated. #thetaphichi alpha beta gamma Rf Rf/sigma Sol_RF 1 0.000.000.000.000.000.00 870.5 21.59 Sol_RF 258.61 -10.17 180.00 169.83 -117.23 10.17 162.5 4.03 Sol_RF 366.02 -0.00 180.00 180.00 -132.030.00 161.1 4.00 Sol_RF 458.42 -9.54 180.00 170.46 -116.859.54 159.8 3.96 Sol_RF 5 149.840.00 180.00 -180.00 60.320.00 156.0 3.87 Sol_RF 658.96 -5.52 180.00 174.48 -117.915.52 151.5 3.76 Sol_RF 765.59 20.95 180.00 20.95 131.18 159.05 143.9 3.57 Sol_RF 890.00 -98.96 180.000.00 180.00 17.92 142.9 3.55 Sol_RF 956.53 15.78 180.00 15.78 113.07 164.22 142.0 3.52 Sol_RF 1071.10 -19.94 180.00 160.06 -142.20 19.94 141.6 3.51 Sol_RF 1171.28 29.78 180.00 29.78 142.55 150.22 140.4 3.48 Sol_RF 1265.22 -15.88 180.00 164.12 -130.44 15.88 139.2 3.45 Sol_RF 1368.84 -0.00 180.00 180.00 -137.670.00 138.0 3.42 Sol_RF 1432.51 -180.00 180.00 -180.00 65.02 -0.00 137.9 3.42 Sol_RF 1575.02 -28.84 180.00 151.16 -150.04 28.84 134.7 3.34 Sol_RF 1671.69 -20.99 180.00 159.01 -143.37 20.99 133.0 3.30 Sol_RF 1792.13 101.46 179.93 102.35 -175.74 79.42 130.9 3.25 Sol_RF 18 107.89 144.73 179.79 145.06 -144.22 35.61 128.8 3.19 Sol_RF 1987.45 -78.19 180.00 101.81 -174.90 78.19 128.1 3.18 Sol_RF 2038.570.69 30.36 102.66 -18.79 -78.71 122.4 3.04 Sol_RF 2126.77 174.59 176.58 172.68 53.523.49 120.5 2.99 Sol_RF 22 116.66 178.08 175.143.49 126.48 187.32 120.5 2.99 Sol_RF 2375.56 -41.35 180.00 138.65 -151.12 41.35 119.8 2.97 Sol_RF 2466.12 36.35 180.00 36.35 132.24 143.65 116.6 2.89 Sol_RF 2583.87 71.62 180.00 71.62 167.74 108.38 114.7 2.85 Sol_RF 2669.24 -12.37 180.00 167.63 -138.48 12.37 112.3 2.79 Sol_RF 2759.75 15.26 172.297.64 119.07 157.12 112.2 2.78 Sol_RF 28 120.25 -164.74 172.29 22.88 119.07 172.36 112.2 2.78 Sol_RF 2996.68 -70.99 180.00 109.01 -166.63 70.99 110.9 2.75 Sol_RF 3063.23 -44.73 180.00 135.27 -126.47 44.73 108.9 2.70 Cheers, Paul Holland -- Yu Zhang HHMI associate Waksman Institute Rutgers University
Re: [ccp4bb] Effect of NCS on estimate of data:parameter ratio
I think you could answer this by performing the following thought experiment: 1. Refine the structure to convergence using strict NCS constraints. 2. Switch to using the equivalent 'infinite-in-the-limit weight' restraints, keeping everything else as is continue refinement of the output from step 1 (assume limitations of finite precision in the code have been overcome by re-programming using whatever precision arithmetic is necessary). 3. Does there exist some finite value of the weight such that the structure changes by less than the experimental error (say by 0.2 Ang RMSD) at step 2 (and if not then why not?). If the answer to (3) is yes, then there's no significant difference in effect between constraints 'infinite weight' restraints. Cheers -- Ian On Thu, Sep 23, 2010 at 5:06 PM, MARTYN SYMMONS martainn_oshioma...@btinternet.com wrote: Dear All one thing I remembered from what Gerard pointed out was the difference in the XPLOR/CNS formalism between strict and restrained which is not a continuum. Restrained was obviously when you had multiple copies and they were restrained with a weight (which was like a force constant) to be similar when superimposed. So if you increase the force constant then they can move during refinement but they all try to move together when they move. And the other extreme is strict where there was no force applied at all but only a single copy of the chain and the ASU is built by applying the NCS symmetry. The atoms are free to move but, unlike the case with restrained where there is superimposition on the fly, in the strict case there is no automatic update of the superimposition matrices. So every move gets religiously copied to all the chains when the ASU is made. At this point I guess the copies can bump and so apply a force on each other but that is a local, and likely to be perturbing, force. best wishes Martyn Martyn Symmons Cambridge --- On Thu, 23/9/10, Ian Tickle ianj...@gmail.com wrote: From: Ian Tickle ianj...@gmail.com Subject: Re: [ccp4bb] Effect of NCS on estimate of data:parameter ratio To: CCP4BB@JISCMAIL.AC.UK Date: Thursday, 23 September, 2010, 11:21 Hi Gerard Pavel Isn't this the proviso I was referring to, that one cannot in practice use an infinite weight because of rounding errors in the target function. The weight just has to be 'big enough' such that the restraint residual becomes sufficiently small that it's no longer significant. In numerical constrained optimisation the method of increasing the constraint weights (a.k.a. 'penalty coefficients') until the constraint violations are sufficiently small is called the 'penalty method', see http://en.wikipedia.org/wiki/Penalty_method . The method where you substitute some of the parameters using the constraint equations is called (you guessed it!) the 'substitution method', see http://people.ucsc.edu/~rgil/Optimization.pdf . There are several other methods, e.g. the 'augmented Lagrangian method' is very popular, see http://www.ualberta.ca/CNS/RESEARCH/NAG/FastfloDoc/Tutorial/html/node112.html . As in the penalty method, the AL method adds additional parameters to be determined (the Lagrange multipliers, one per constraint) instead of eliminating some parameters using the constraint equations; however the advantage is that it removes the requirement that the penalty coefficient be very big. The point about all these methods of constrained optimisation is that they are in principle only different ways of achieving the same result, at least that's what the textbooks say! And now after the penalties and substitutions it's time to blow the whistle ... Cheers -- Ian On Wed, Sep 22, 2010 at 10:00 PM, Pavel Afonine pafon...@lbl.gov wrote: I agree with Gerard. Example: it's unlikely to achieve a result of rigid-body refinement (when you refine six rotation/translation parameters) by replacing it with refining individual coordinates using infinitely large weights for restraints. Pavel. On 9/22/10 1:46 PM, Gerard DVD Kleywegt wrote: Hi Ian, First, constraints are just a special case of restraints in the limit of infinite weights, in fact one way of getting constraints is simply to use restraints with very large weights (though not too large that you get rounding problems). These 'pseudo-constraints' will be indistinguishable in effect from the 'real thing'. So why treat restraints and constraints differently as far as the statistics are concerned: the difference is purely one of implementation. In practice this is not true, of course. If you impose infinitely strong NCS restraints, any change to a thusly restrained parameter by the refinement program will make the target function infinite, so effectively your model will never change. This is very different from the behaviour under NCS constraints and the resulting models in these two
