Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight, that they have been unknowingly achieving super-resolution all the time, from the grand old days of Bob Diamond's real-space refinement program - just like Monsieur Jourdain found out that he had been speaking in prose all his life without realising it. I guess that super-resolution is a sexier keyword in the mind of editors of Nature that restrained crystallographic refinement :-)) ! With best wishes for the New Year, Gerard. -- We found DEN to be useful to move some atoms into correct positions in cases where electron density maps are difficult or impossible to interpret at low resolution. By default, DEN is active during the first torsion angle molecular dynamics stages, but then turned off during the last two stages. In addition, the DEN network is deformable. Thus, DEN is very different from secondary structure restraints or point restraints to reference models which are on all the time. Rather, DEN steers or guides the torsion angle conformational search process during refinement. Cheers, Axel On Dec 24, 2010, at 2:14 PM, Bernhard Rupp (Hofkristallrat a.D.) wrote: I find the super-resolution claims in this paper a bit of a conjuring trick. I think it is understood that information cannot come from nothing. You cannot cheat in basic physics. Interestingly, I had the same discussion with bioinformatics colleagues a short time ago. The problem is the same and seems of a semantic nature. They are using prior information of some sort (undisclosed) to successfully improve maps and they suggested to call this 'resolution increase'. I had the same objection and said that in crystallography resolution is a relatively hard term defined by the degree to which experimental observations are available, and as crystallographers we won't like that claim at all. On the other side it is uncontested that as long as the model fits (crossvalidation-) data better when prior information is used, something useful has been achieved - again with all the caveats of weights and bias etc admitted. However, how to entice non-experts to actually use new methods is another thing, and here the semantics come in. In essence, if at the end it results in better structures, how much of the unfortunately but undeniably necessary salesmanship is just right or acceptable? Within contemporary social constraints (aka Zeitgeist) that remains pretty much an infinitely debatable matter.. Merry Christmas, BR -- Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard. Axel T. Brunger Investigator, Howard Hughes Medical Institute Professor of Molecular and Cellular Physiology Stanford University Web:http://atbweb.stanford.edu Email: brun...@stanford.edu Phone: +1 650-736-1031 Fax:+1 650-745-1463 -- === * * * Gerard Bricogne g...@globalphasing.com * * * * Global Phasing Ltd. * * Sheraton House, Castle Park Tel: +44-(0)1223-353033 * * Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 * *
[ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Begin forwarded message: From: Charles W. Carter, Jr car...@med.unc.edu Date: January 6, 2011 10:51:20 AM GMT+01:00 To: Gerard Bricogne g...@globalphasing.com Subject: Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies I echo Gérard's thought. Pascal Retailleau did a relevant experiment published in Acta D: Retailleau, et al., (2001) High-resolution experimental phases for tryptophanyl-tRNA synthetase (TrpRS) complexed with tryptophanyl-5'AMP, Acta Cryst, D57, 1595–1608 He determined three independent sets of experimental phases for two different 1.7 Å selenomethionine structures (SAD) plus a 1.6 Å native (MIRAS) and refined the structures independently. The rmsd between the two SeMet structures was 0.25 Å, whereas that between the two SAD structures and the native structure was 0.39 Å, sufficient to demonstrate significant differences between the SeMet and native proteins. This experimental variance is a quite considerable indication of the magnitude of coordinate errors. Thus, as Gérard, who also was an author on that work together with Bob Sweet, points out, we're delighted to discover we have been achieving super-resolution to use Axel's neologism! Charlie On Jan 6, 2011, at 10:13 AM, Gerard Bricogne wrote: Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight, that they have been unknowingly achieving super-resolution all the time, from the grand old days of Bob Diamond's real-space refinement program - just like Monsieur Jourdain found out that he had been speaking in prose all his life without realising it. I guess that super-resolution is a sexier keyword in the mind of editors of Nature that restrained crystallographic refinement :-)) ! With best wishes for the New Year, Gerard. -- We found DEN to be useful to move some atoms into correct positions in cases where electron density maps are difficult or impossible to interpret at low resolution. By default, DEN is active during the first torsion angle molecular dynamics stages, but then turned off during the last two stages. In addition, the DEN network is deformable. Thus, DEN is very different from secondary structure restraints or point restraints to reference models which are on all the time. Rather, DEN steers or guides the torsion angle conformational search process during refinement. Cheers, Axel On Dec 24, 2010, at 2:14 PM, Bernhard Rupp (Hofkristallrat a.D.) wrote: I find the super-resolution claims in this paper a bit of a conjuring trick. I think it is understood that information cannot come from nothing. You cannot cheat in basic physics. Interestingly, I had the same discussion with bioinformatics colleagues a short time ago. The problem is the same and seems of a semantic nature. They are using prior information of some sort (undisclosed) to successfully improve maps and they suggested to call this 'resolution increase'. I had the same objection and said that in crystallography resolution is a relatively hard term defined by the degree to which experimental observations are available, and as crystallographers we won't like that claim at all. On the other side it is uncontested that as long as the model fits (crossvalidation-) data better when prior information is used, something useful has been achieved - again with all the caveats of weights and bias etc admitted. However, how to entice non-experts to actually use new methods is another thing, and here the semantics come in. In essence, if at the end it results in better structures, how much of the unfortunately but undeniably necessary salesmanship is just right or acceptable? Within contemporary social constraints (aka Zeitgeist) that remains pretty much an infinitely debatable matter.. Merry Christmas, BR -- Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure
Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
I too think the phrase super-resolution is rather misleading, in particular the analogy with light microscopy methods. Super-resolution in these latter cases is achieved via different physical phenomena (think excitations not waves). Would one claim super-resolution when refining the relative positions of the carbon atoms in benzene given the constraints of 6 fold symmetry and a carbon carbon distance of 1.39 angstroms? What would Moliere think? However, to quote from the DEN paper Our approach is a major advance over conventional modeling of low resolution X-ray diffraction data by fitting rigid bodies since it accounts for deformations of the models while at the same time using a minimal set of variables (the single-bond torsion angles) Overall this seems a reasonable claim. Colin From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Charles W. Carter, Jr Sent: 06 January 2011 09:52 To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies Begin forwarded message: From: Charles W. Carter, Jr car...@med.unc.edu Date: January 6, 2011 10:51:20 AM GMT+01:00 To: Gerard Bricogne g...@globalphasing.com Subject: Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies I echo Gérard's thought. Pascal Retailleau did a relevant experiment published in Acta D: Retailleau, et al., (2001) High-resolution experimental phases for tryptophanyl-tRNA synthetase (TrpRS) complexed with tryptophanyl-5'AMP, Acta Cryst, D57, 1595-1608 He determined three independent sets of experimental phases for two different 1.7 Å selenomethionine structures (SAD) plus a 1.6 Å native (MIRAS) and refined the structures independently. The rmsd between the two SeMet structures was 0.25 Å, whereas that between the two SAD structures and the native structure was 0.39 Å, sufficient to demonstrate significant differences between the SeMet and native proteins. This experimental variance is a quite considerable indication of the magnitude of coordinate errors. Thus, as Gérard, who also was an author on that work together with Bob Sweet, points out, we're delighted to discover we have been achieving super-resolution to use Axel's neologism! Charlie On Jan 6, 2011, at 10:13 AM, Gerard Bricogne wrote: Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight, that they have been unknowingly achieving super-resolution all the time, from the grand old days of Bob Diamond's real-space refinement program - just like Monsieur Jourdain found out that he had been speaking in prose all his life without realising it. I guess that super-resolution is a sexier keyword in the mind of editors of Nature that restrained crystallographic refinement :-)) ! With best wishes for the New Year, Gerard. -- We found DEN to be useful to move some atoms into correct positions in cases where electron density maps are difficult or impossible to interpret at low resolution. By default, DEN is active during the first torsion angle molecular dynamics stages, but then turned off during the last two stages. In addition, the DEN network is deformable. Thus, DEN is very different from secondary structure restraints or point restraints to reference models which are on all the time. Rather, DEN steers or guides the torsion angle conformational search process during refinement. Cheers, Axel On Dec 24, 2010, at 2:14 PM, Bernhard Rupp (Hofkristallrat a.D.) wrote: I find the super-resolution claims in this paper a bit of a conjuring trick. I think it is understood that information cannot come from nothing. You cannot cheat in basic physics. Interestingly, I had the same discussion with bioinformatics colleagues a short time ago. The problem is the same and seems of a semantic nature. They are using prior information of some sort (undisclosed) to successfully improve maps
Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Maybe if you are really into this, you could go public by writing a letter to the editor (from the CCP4BB?), and I am sure there will be a good response from the authors... Jacob *** Jacob Pearson Keller Northwestern University Medical Scientist Training Program cel: 773.608.9185 email: j-kell...@northwestern.edu ***
Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Dear Colin, Wladek Minor has just drawn my attention to the following recent paper: Acta Cryst. (2010). D66, 1041.1042 (that I must admit to having failed to notice) also expressing reservations about some uses of creative language. With best wishes, Gerard. -- On Thu, Jan 06, 2011 at 11:13:41AM -, Colin Nave wrote: I too think the phrase super-resolution is rather misleading, in particular the analogy with light microscopy methods. Super-resolution in these latter cases is achieved via different physical phenomena (think excitations not waves). Would one claim super-resolution when refining the relative positions of the carbon atoms in benzene given the constraints of 6 fold symmetry and a carbon carbon distance of 1.39 angstroms? What would Moliere think? However, to quote from the DEN paper Our approach is a major advance over conventional modeling of low resolution X-ray diffraction data by fitting rigid bodies since it accounts for deformations of the models while at the same time using a minimal set of variables (the single-bond torsion angles) Overall this seems a reasonable claim. Colin From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Charles W. Carter, Jr Sent: 06 January 2011 09:52 To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies Begin forwarded message: From: Charles W. Carter, Jr car...@med.unc.edu Date: January 6, 2011 10:51:20 AM GMT+01:00 To: Gerard Bricogne g...@globalphasing.com Subject: Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies I echo Gérard's thought. Pascal Retailleau did a relevant experiment published in Acta D: Retailleau, et al., (2001) High-resolution experimental phases for tryptophanyl-tRNA synthetase (TrpRS) complexed with tryptophanyl-5'AMP, Acta Cryst, D57, 1595-1608 He determined three independent sets of experimental phases for two different 1.7 Å selenomethionine structures (SAD) plus a 1.6 Å native (MIRAS) and refined the structures independently. The rmsd between the two SeMet structures was 0.25 Å, whereas that between the two SAD structures and the native structure was 0.39 Å, sufficient to demonstrate significant differences between the SeMet and native proteins. This experimental variance is a quite considerable indication of the magnitude of coordinate errors. Thus, as Gérard, who also was an author on that work together with Bob Sweet, points out, we're delighted to discover we have been achieving super-resolution to use Axel's neologism! Charlie On Jan 6, 2011, at 10:13 AM, Gerard Bricogne wrote: Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight, that they have been unknowingly achieving super-resolution all the time, from the grand old days of Bob Diamond's real-space refinement program - just like Monsieur Jourdain found out that he had been speaking in prose all his life without realising it. I guess that super-resolution is a sexier keyword in the mind of editors of Nature that restrained crystallographic refinement :-)) ! With best wishes for the New Year, Gerard. -- We found DEN to be useful to move some atoms into correct positions in cases where electron density maps are difficult or impossible to interpret at low resolution. By default, DEN is active during the first torsion angle molecular dynamics stages, but then turned off during the last two stages. In addition, the DEN network is deformable. Thus, DEN is very different from secondary structure restraints or point restraints to reference models which are on all the time. Rather, DEN steers or guides the torsion angle conformational search process during refinement. Cheers, Axel On Dec 24, 2010, at 2:14 PM, Bernhard Rupp (Hofkristallrat a.D.) wrote: I find the super-resolution
Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Hi, creative language, you say... ha-ha. Go through this: A number of publications define ultrahigh or subatomic resolution in the range of 1.0–0.5A (Lecomte et al., 2008; Petrova et al., 2006; Howard et al., 2004; Guillot et al., 2008; Housset et al., 2000). Right? When you see ultra-high resolution, you can arbitrary pick any number from 1 to 0.5A, and call it ultra-high resolution, just arbitrarily. What happens at lower resolution end is somewhat a bigger mess simply because there is no such eye-catching names, so super-resolution looks just fine to me - just to fill the gap, unless you know which numbers you mean. And to know these numbers you probably need to carefully read this paper. Not necessarily it will give you the numbers, but definitely the ideas to think about: Acta Cryst. (2009). D65, 1283-1291 On the use of logarithmic scales for analysis of diffraction data A. Urzhumtsev, P. V. Afonine and P. D. Adams Good luck! Pavel. On Thu, Jan 6, 2011 at 2:48 PM, Gerard Bricogne g...@globalphasing.comwrote: Dear Colin, Wladek Minor has just drawn my attention to the following recent paper: Acta Cryst. (2010). D66, 1041.1042 (that I must admit to having failed to notice) also expressing reservations about some uses of creative language. With best wishes, Gerard. -- On Thu, Jan 06, 2011 at 11:13:41AM -, Colin Nave wrote: I too think the phrase super-resolution is rather misleading, in particular the analogy with light microscopy methods. Super-resolution in these latter cases is achieved via different physical phenomena (think excitations not waves). Would one claim super-resolution when refining the relative positions of the carbon atoms in benzene given the constraints of 6 fold symmetry and a carbon carbon distance of 1.39 angstroms? What would Moliere think? However, to quote from the DEN paper Our approach is a major advance over conventional modeling of low resolution X-ray diffraction data by fitting rigid bodies since it accounts for deformations of the models while at the same time using a minimal set of variables (the single-bond torsion angles) Overall this seems a reasonable claim. Colin From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Charles W. Carter, Jr Sent: 06 January 2011 09:52 To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies Begin forwarded message: From: Charles W. Carter, Jr car...@med.unc.edu Date: January 6, 2011 10:51:20 AM GMT+01:00 To: Gerard Bricogne g...@globalphasing.com Subject: Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies I echo Gérard's thought. Pascal Retailleau did a relevant experiment published in Acta D: Retailleau, et al., (2001) High-resolution experimental phases for tryptophanyl-tRNA synthetase (TrpRS) complexed with tryptophanyl-5'AMP, Acta Cryst, D57, 1595-1608 He determined three independent sets of experimental phases for two different 1.7 Å selenomethionine structures (SAD) plus a 1.6 Å native (MIRAS) and refined the structures independently. The rmsd between the two SeMet structures was 0.25 Å, whereas that between the two SAD structures and the native structure was 0.39 Å, sufficient to demonstrate significant differences between the SeMet and native proteins. This experimental variance is a quite considerable indication of the magnitude of coordinate errors. Thus, as Gérard, who also was an author on that work together with Bob Sweet, points out, we're delighted to discover we have been achieving super-resolution to use Axel's neologism! Charlie On Jan 6, 2011, at 10:13 AM, Gerard Bricogne wrote: Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight, that they have been unknowingly achieving super-resolution all the time, from the grand old days of Bob Diamond's real-space refinement program - just like Monsieur Jourdain found out that he had been speaking in prose all his life without realising it. I guess that super-resolution is a sexier keyword in the mind
Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Dont forget all the atomic resolution 3Å structures! From: Pavel Afonine [pafon...@gmail.com] Sent: Thursday, January 06, 2011 6:23 PM To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies Hi, creative language, you say... ha-ha. Go through this: A number of publications define ultrahigh or subatomic resolution in the range of 1.0–0.5A (Lecomte et al., 2008; Petrova et al., 2006; Howard et al., 2004; Guillot et al., 2008; Housset et al., 2000). Right? When you see ultra-high resolution, you can arbitrary pick any number from 1 to 0.5A, and call it ultra-high resolution, just arbitrarily. What happens at lower resolution end is somewhat a bigger mess simply because there is no such eye-catching names, so super-resolution looks just fine to me - just to fill the gap, unless you know which numbers you mean. And to know these numbers you probably need to carefully read this paper. Not necessarily it will give you the numbers, but definitely the ideas to think about: Acta Cryst. (2009). D65, 1283-1291 On the use of logarithmic scales for analysis of diffraction data A. Urzhumtsev, P. V. Afonine and P. D. Adams Good luck! Pavel. On Thu, Jan 6, 2011 at 2:48 PM, Gerard Bricogne g...@globalphasing.commailto:g...@globalphasing.com wrote: Dear Colin, Wladek Minor has just drawn my attention to the following recent paper: Acta Cryst. (2010). D66, 1041.1042 (that I must admit to having failed to notice) also expressing reservations about some uses of creative language. With best wishes, Gerard. -- On Thu, Jan 06, 2011 at 11:13:41AM -, Colin Nave wrote: I too think the phrase super-resolution is rather misleading, in particular the analogy with light microscopy methods. Super-resolution in these latter cases is achieved via different physical phenomena (think excitations not waves). Would one claim super-resolution when refining the relative positions of the carbon atoms in benzene given the constraints of 6 fold symmetry and a carbon carbon distance of 1.39 angstroms? What would Moliere think? However, to quote from the DEN paper Our approach is a major advance over conventional modeling of low resolution X-ray diffraction data by fitting rigid bodies since it accounts for deformations of the models while at the same time using a minimal set of variables (the single-bond torsion angles) Overall this seems a reasonable claim. Colin From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UKmailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Charles W. Carter, Jr Sent: 06 January 2011 09:52 To: CCP4BB@JISCMAIL.AC.UKmailto:CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Fwd: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies Begin forwarded message: From: Charles W. Carter, Jr car...@med.unc.edumailto:car...@med.unc.edu Date: January 6, 2011 10:51:20 AM GMT+01:00 To: Gerard Bricogne g...@globalphasing.commailto:g...@globalphasing.com Subject: Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies I echo Gérard's thought. Pascal Retailleau did a relevant experiment published in Acta D: Retailleau, et al., (2001) High-resolution experimental phases for tryptophanyl-tRNA synthetase (TrpRS) complexed with tryptophanyl-5'AMP, Acta Cryst, D57, 1595-1608 He determined three independent sets of experimental phases for two different 1.7 Å selenomethionine structures (SAD) plus a 1.6 Å native (MIRAS) and refined the structures independently. The rmsd between the two SeMet structures was 0.25 Å, whereas that between the two SAD structures and the native structure was 0.39 Å, sufficient to demonstrate significant differences between the SeMet and native proteins. This experimental variance is a quite considerable indication of the magnitude of coordinate errors. Thus, as Gérard, who also was an author on that work together with Bob Sweet, points out, we're delighted to discover we have been achieving super-resolution to use Axel's neologism! Charlie On Jan 6, 2011, at 10:13 AM, Gerard Bricogne wrote: Dear Axel, On Sun, Dec 26, 2010 at 01:15:44PM -0800, Axel Brunger wrote: We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). In that case, all crystallographers doing stereochemically restrained refinement will now have become aware, to their great delight
Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
We defined super-resolution in our DEN paper as achieving coordinate accuracy better than the resolution limit d_min of the diffraction data. We proposed this definition in analogy to its use wide-spread use in optical microscopy: super-resolution methods such as STORM, PALM, and STED achieve accuracy of positions of fluorescent labels significantly better than the diffraction limit (in some cases, sub-nanometer accuracy - Pertsinidis, Zhang, Chu, Nature 466, 647-651, 2010). We found DEN to be useful to move some atoms into correct positions in cases where electron density maps are difficult or impossible to interpret at low resolution. By default, DEN is active during the first torsion angle molecular dynamics stages, but then turned off during the last two stages. In addition, the DEN network is deformable. Thus, DEN is very different from secondary structure restraints or point restraints to reference models which are on all the time. Rather, DEN steers or guides the torsion angle conformational search process during refinement. Cheers, Axel On Dec 24, 2010, at 2:14 PM, Bernhard Rupp (Hofkristallrat a.D.) wrote: I find the super-resolution claims in this paper a bit of a conjuring trick. I think it is understood that information cannot come from nothing. You cannot cheat in basic physics. Interestingly, I had the same discussion with bioinformatics colleagues a short time ago. The problem is the same and seems of a semantic nature. They are using prior information of some sort (undisclosed) to successfully improve maps and they suggested to call this 'resolution increase'. I had the same objection and said that in crystallography resolution is a relatively hard term defined by the degree to which experimental observations are available, and as crystallographers we won't like that claim at all. On the other side it is uncontested that as long as the model fits (crossvalidation-) data better when prior information is used, something useful has been achieved - again with all the caveats of weights and bias etc admitted. However, how to entice non-experts to actually use new methods is another thing, and here the semantics come in. In essence, if at the end it results in better structures, how much of the unfortunately but undeniably necessary salesmanship is just right or acceptable? Within contemporary social constraints (aka Zeitgeist) that remains pretty much an infinitely debatable matter.. Merry Christmas, BR -- Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard. Axel T. Brunger Investigator, Howard Hughes Medical Institute Professor of Molecular and Cellular Physiology Stanford University Web:http://atbweb.stanford.edu Email: brun...@stanford.edu Phone: +1 650-736-1031 Fax:+1 650-745-1463
Re: [ccp4bb] Resolution and distance accuracies
I have a program which computes the atomic electron density profile (attached) as you would see it in a map, using accurate scattering factors and taking the resolution limit into account. I wouldn't call the profile for a C atom with B=100 at 2.5 Ang resolution 'flat', maybe 'flatter'. 'Flat' would imply that it's lost in the noise of other atoms with B=100. My point is that it's relative. Since my average B is 85 Ang.^2, an individual B of 100 or even 120 doesn't seem out of the ordinary at all. If the average B were 10 then I would agree that anything over say 50 would appear flat and insignificant. The reason I think is simply that atoms with low B factor have series termination ripples around them which can swamp the density of other atoms with high B factor (for example the ripples from a B=10 C atom are half the height of the peak of a B=150 atom). So the net 'noise' level in a map with low average B is much higher than in one with high average B, so that any atoms with high individual B just get lost in the noise. Cheers -- Ian On Thu, Dec 23, 2010 at 8:05 PM, Ronald E Stenkamp stenk...@u.washington.edu wrote: Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR attachment: c1.png
Re: [ccp4bb] Resolution and distance accuracies
To get an idea about this signal:noise issue, I wonder what the B factor of bulk solvent would be, if it is possible to think in such terms, i.e., fill the bulk solvent with dummy waters and let the B's do what they will. I wonder whether anyone has tried explicitly modelling the whole unit cell this way, and what the effects were? I imagine it might be possible at ultra-high resolution. In other words, I am wondering whether the electron density for a 100-B-factor atom is significantly different from bulk solvent. Jacob On Fri, Dec 24, 2010 at 6:26 AM, Ian Tickle ianj...@gmail.com wrote: I have a program which computes the atomic electron density profile (attached) as you would see it in a map, using accurate scattering factors and taking the resolution limit into account. I wouldn't call the profile for a C atom with B=100 at 2.5 Ang resolution 'flat', maybe 'flatter'. 'Flat' would imply that it's lost in the noise of other atoms with B=100. My point is that it's relative. Since my average B is 85 Ang.^2, an individual B of 100 or even 120 doesn't seem out of the ordinary at all. If the average B were 10 then I would agree that anything over say 50 would appear flat and insignificant. The reason I think is simply that atoms with low B factor have series termination ripples around them which can swamp the density of other atoms with high B factor (for example the ripples from a B=10 C atom are half the height of the peak of a B=150 atom). So the net 'noise' level in a map with low average B is much higher than in one with high average B, so that any atoms with high individual B just get lost in the noise. Cheers -- Ian On Thu, Dec 23, 2010 at 8:05 PM, Ronald E Stenkamp stenk...@u.washington.edu wrote: Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR -- *** Jacob Pearson Keller Northwestern University Medical Scientist Training Program cel: 773.608.9185 email: j-kell...@northwestern.edu ***
Re: [ccp4bb] Resolution and distance accuracies
In reference to the low and high B atoms in a map and the ripples, one can actually calculate something of educational value using my web applets (with all the caveats imposed by simplicity of the 1-d case). Here is how: Go to http://www.ruppweb.org/new_comp/structure_factors.htm and set the B-value of the C atom to 100 (yes it takes integer 100 in contrast to instructions) and the B-value of the O atom to 2. Leave the default for the rest and execute (you can pick a name for the SF file if you like). then goto http://www.ruppweb.org/new_comp/fourier_maps.htm set the grid to 100 and optionally enter the file name you used before and execute Look at the resulting map and the peak shapes. The relative scale, broadening, and ripples all show up as discussed. Merry Christmas, BR -Original Message- From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Ian Tickle Sent: Friday, December 24, 2010 4:26 AM To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] Resolution and distance accuracies I have a program which computes the atomic electron density profile (attached) as you would see it in a map, using accurate scattering factors and taking the resolution limit into account. I wouldn't call the profile for a C atom with B=100 at 2.5 Ang resolution 'flat', maybe 'flatter'. 'Flat' would imply that it's lost in the noise of other atoms with B=100. My point is that it's relative. Since my average B is 85 Ang.^2, an individual B of 100 or even 120 doesn't seem out of the ordinary at all. If the average B were 10 then I would agree that anything over say 50 would appear flat and insignificant. The reason I think is simply that atoms with low B factor have series termination ripples around them which can swamp the density of other atoms with high B factor (for example the ripples from a B=10 C atom are half the height of the peak of a B=150 atom). So the net 'noise' level in a map with low average B is much higher than in one with high average B, so that any atoms with high individual B just get lost in the noise. Cheers -- Ian On Thu, Dec 23, 2010 at 8:05 PM, Ronald E Stenkamp stenk...@u.washington.edu wrote: Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR
Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
I find the super-resolution claims in this paper a bit of a conjuring trick. I think it is understood that information cannot come from nothing. You cannot cheat in basic physics. Interestingly, I had the same discussion with bioinformatics colleagues a short time ago. The problem is the same and seems of a semantic nature. They are using prior information of some sort (undisclosed) to successfully improve maps and they suggested to call this 'resolution increase'. I had the same objection and said that in crystallography resolution is a relatively hard term defined by the degree to which experimental observations are available, and as crystallographers we won't like that claim at all. On the other side it is uncontested that as long as the model fits (crossvalidation-) data better when prior information is used, something useful has been achieved - again with all the caveats of weights and bias etc admitted. However, how to entice non-experts to actually use new methods is another thing, and here the semantics come in. In essence, if at the end it results in better structures, how much of the unfortunately but undeniably necessary salesmanship is just right or acceptable? Within contemporary social constraints (aka Zeitgeist) that remains pretty much an infinitely debatable matter.. Merry Christmas, BR -- Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard.
Re: [ccp4bb] Resolution and distance accuracies
Here is another set of relavant papers and instructive pictures: 1) CCP4 Newsletter http://www.ccp4.ac.uk/newsletters/newsletter42/content.html On the Fourier series truncation peaks at subatomic resolution Anne Bochow, Alexandre Urzhumtsev 2) Some Facts About Maps (pages 30, 32, 33): http://www.phenix-online.org/presentations/latest/pavel_maps.pdf 3) Page 267 Figure 4: On the possibility of the observation of valence electron density for individual bonds in proteins in conventional difference maps P. V. Afonine, V. Y. Lunin, N. Muzet and A. Urzhumtsev Acta Cryst. (2004). D60, 260-274 4) Central Ligand in the FeMo-Cofactor Nitrogenase MoFe-Protein at 1.16 Å Resolution: A. Oliver Einsle, et al. Science, 1696 (2002) 297 5) Numerous pictures of density as function of distance from atom center for various B-factors: Acta Cryst. (2004). A60, 19-32 On a fast calculation of structure factors at a subatomic resolution P. V. Afonine and A. Urzhumtsev Pavel. On Fri, Dec 24, 2010 at 1:31 PM, Bernhard Rupp (Hofkristallrat a.D.) hofkristall...@gmail.com wrote: In reference to the low and high B atoms in a map and the ripples, one can actually calculate something of educational value using my web applets (with all the caveats imposed by simplicity of the 1-d case). Here is how: Go to http://www.ruppweb.org/new_comp/structure_factors.htm and set the B-value of the C atom to 100 (yes it takes integer 100 in contrast to instructions) and the B-value of the O atom to 2. Leave the default for the rest and execute (you can pick a name for the SF file if you like). then goto http://www.ruppweb.org/new_comp/fourier_maps.htm set the grid to 100 and optionally enter the file name you used before and execute Look at the resulting map and the peak shapes. The relative scale, broadening, and ripples all show up as discussed. Merry Christmas, BR -Original Message- From: CCP4 bulletin board [mailto:ccp...@jiscmail.ac.uk] On Behalf Of Ian Tickle Sent: Friday, December 24, 2010 4:26 AM To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] Resolution and distance accuracies I have a program which computes the atomic electron density profile (attached) as you would see it in a map, using accurate scattering factors and taking the resolution limit into account. I wouldn't call the profile for a C atom with B=100 at 2.5 Ang resolution 'flat', maybe 'flatter'. 'Flat' would imply that it's lost in the noise of other atoms with B=100. My point is that it's relative. Since my average B is 85 Ang.^2, an individual B of 100 or even 120 doesn't seem out of the ordinary at all. If the average B were 10 then I would agree that anything over say 50 would appear flat and insignificant. The reason I think is simply that atoms with low B factor have series termination ripples around them which can swamp the density of other atoms with high B factor (for example the ripples from a B=10 C atom are half the height of the peak of a B=150 atom). So the net 'noise' level in a map with low average B is much higher than in one with high average B, so that any atoms with high individual B just get lost in the noise. Cheers -- Ian On Thu, Dec 23, 2010 at 8:05 PM, Ronald E Stenkamp stenk...@u.washington.edu wrote: Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR
[ccp4bb] Resolution and distance accuracies
We clearly have confidence in distance measurements in crystal structures of an order of magnitude better than the resolution ie 0.1-0.3 Angstroms, but can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Have a Happy Christmas and see many of you at CCP4 Nick
Re: [ccp4bb] Resolution and distance accuracies
Dear Nick I think the point here is more the precision achieved via the fitting of atom positions according to the gradient of the electron density. There is obviously an overall link of the detail of/in this gradient to the measurable diffraction resolution, which can be estimated in a number of ways. This apparently simple statement of course excludes any systematic effects of data incompleteness that may exist in a data set or variations accruing from different atomic B factors. The Cruickshank Diffraction Precision Index, and especially the David Blow reformulation into explicit experimental parameters, is a powerful overall descriptor in my view. Seasons greetings, John On Thu, Dec 23, 2010 at 9:20 AM, Nicholas keep n.k...@mail.cryst.bbk.ac.uk wrote: We clearly have confidence in distance measurements in crystal structures of an order of magnitude better than the resolution ie 0.1-0.3 Angstroms, but can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Have a Happy Christmas and see many of you at CCP4 Nick -- Professor John R Helliwell DSc
Re: [ccp4bb] Resolution and distance accuracies
Hi Nick The 0.1 to 0.3 Ang range you quote obviously only applies to the kind of atoms we as macromolecular crystallographers are interested in, i.e. mostly carbon with some nitrogen oxygen, and also at the 'typical' data resolution and B factors that we observe. For hydrogen it would obviously be a lot more, for heavier atoms a good deal less. My point is that the positional uncertainty (I assume that by 'distance' you meant 'position' not 'bond length', otherwise we also have to consider the effect of geometric restraints) is strongly dependent on the atomic number and this indeed gives the clue as to its origin. The prior information about _atomicity_ that we are adding (via the scattering factors) exerts a powerful effect in reducing the positional uncertainty well below the nominal resolution, just as prior information in the form of geometric restraints reduces the uncertainties in the bond lengths and angles. David Moss I (with Roman Laskowski) published something on the lines of correlating positional precision with atomic number, B factor, resolution and geometric restraints while I was at Birkbeck [Acta Cryst. (1998). D54, 243-252], and I know others (Cruickshank, Blow, Sheldrick) have done something similar before and since. Cheers -- Ian On Thu, Dec 23, 2010 at 9:20 AM, Nicholas keep n.k...@mail.cryst.bbk.ac.uk wrote: We clearly have confidence in distance measurements in crystal structures of an order of magnitude better than the resolution ie 0.1-0.3 Angstroms, but can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Have a Happy Christmas and see many of you at CCP4 Nick
Re: [ccp4bb] Resolution and distance accuracies
This posting concerning positional precision prompted me to recall a CCP4 meeting (not a recent one) where a well-known crystallographer (who shall remain nameless - present company excepted BTW - AFAIK the crystallographer in question has never posted to this BB) claimed that an atom with a B factor of 100 Ang^2 is undetectable in a map because it has a positional uncertainty of 1 Ang, as given by its RMS vibration amplitude (sqrt(100/8pi^2) to be precise). This is nonsense of course: I have a structure where some atoms have B factors of 120 Ang^2 and their density is indisputable. While it's certainly true that atoms with high B factors tend to have large uncertainties, the calculation is not as simple as that, and other factors (such as the resolution and the atomic number) are involved. As an analogy consider the Foucault's Pendulum in the Panthéon in Paris. This particular pendulum has a maximum displacement of about 2 metres, so an RMS amplitude ~ 1.4 metres. However I would guess that its mean position (i.e. at the centre of a swing directly under the point of suspension) is known to within a few mm. Positional uncertainty is of course the uncertainty in the _average_ position of an atom which is not directly related to its RMS amplitude of motion. -- Ian On Thu, Dec 23, 2010 at 9:20 AM, Nicholas keep n.k...@mail.cryst.bbk.ac.uk wrote: We clearly have confidence in distance measurements in crystal structures of an order of magnitude better than the resolution ie 0.1-0.3 Angstroms, but can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Have a Happy Christmas and see many of you at CCP4 Nick
Re: [ccp4bb] Resolution and distance accuracies
can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR
Re: [ccp4bb] Resolution and distance accuracies
Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR
[ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Oops I am outdated: Axel just emailed me that he describes an improved coordinate estimate beyond the Rayleigh criterion in his recent paper Schroder GF, Levitt M, Brunger AT (2010) Super-resolution biomolecular crystallography with low-resolution data. Nature 464(7292), 1218-1222. For the deformable elastic network (DEN) refinement, see his ref 14 Schroder GF, Brunger AT, Levitt M (2007) Combining efficient conformational sampling with a deformable elastic network model facilitates structure refinement at low resolution. Structure 15(12), 1630-1641. BR
Re: [ccp4bb] Resolution and distance accuracies
electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps Hmmthen there would not be any low resolution structures: Say you have a low resolution structure, 3.5 A with a mean B of ~100A2 or so. Then on average all density peaks are broad and flat (FT of narrow SF=broad ED). If you contour down correspondingly (in absolute terms), that looks then just like a low resolution map. I see no problem there. But you are right insofar as a 100A2 atom in a high resolution map - properly contoured for that resolution - will not show much high level density, consistent with no scattering contribution at high resolution. And at very low density contour levels, the broad low-resolution density of that atom may also be obscured in noise from the remaining high density contributions. Short of contour levels and noise issues, I can't see any contradiction or problem here? Best, BR -Original Message- From: Ronald E Stenkamp [mailto:stenk...@u.washington.edu] Sent: Thursday, December 23, 2010 12:06 PM To: Bernhard Rupp (Hofkristallrat a.D.) Cc: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] Resolution and distance accuracies Something related to the results in the 1984 paper, but never published, is that the calculated electron density for an atom with a B of 100 Angstroms**2 is so flat that you wonder how those atoms can be seen in electron density maps. Ron On Thu, 23 Dec 2010, Bernhard Rupp (Hofkristallrat a.D.) wrote: can anyone point me to a more exact theory of distance accuracy compared to optical resolution, preferably one that would apply to microscopy as well. Stenkamp RE, Jensen LH (1984) Resolution revisited: limit of detail in electron density maps. Acta Crystallogr. A40(3), 251-254. MX, BR
Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard. -- On Thu, Dec 23, 2010 at 12:25:26PM -0800, Bernhard Rupp (Hofkristallrat a.D.) wrote: Oops I am outdated: Axel just emailed me that he describes an improved coordinate estimate beyond the Rayleigh criterion in his recent paper Schroder GF, Levitt M, Brunger AT (2010) Super-resolution biomolecular crystallography with low-resolution data. Nature 464(7292), 1218-1222. For the deformable elastic network (DEN) refinement, see his ref 14 Schroder GF, Brunger AT, Levitt M (2007) Combining efficient conformational sampling with a deformable elastic network model facilitates structure refinement at low resolution. Structure 15(12), 1630-1641. BR -- === * * * Gerard Bricogne g...@globalphasing.com * * * * Global Phasing Ltd. * * Sheraton House, Castle Park Tel: +44-(0)1223-353033 * * Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 * * * ===
Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Extrapolating to the infinite, I wonder how super-resolution fits into MD models or structure predictions? Jacob On Thu, Dec 23, 2010 at 5:04 PM, Gerard Bricogne g...@globalphasing.com wrote: Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard. -- On Thu, Dec 23, 2010 at 12:25:26PM -0800, Bernhard Rupp (Hofkristallrat a.D.) wrote: Oops I am outdated: Axel just emailed me that he describes an improved coordinate estimate beyond the Rayleigh criterion in his recent paper Schroder GF, Levitt M, Brunger AT (2010) Super-resolution biomolecular crystallography with low-resolution data. Nature 464(7292), 1218-1222. For the deformable elastic network (DEN) refinement, see his ref 14 Schroder GF, Brunger AT, Levitt M (2007) Combining efficient conformational sampling with a deformable elastic network model facilitates structure refinement at low resolution. Structure 15(12), 1630-1641. BR -- === * * * Gerard Bricogne g...@globalphasing.com * * * * Global Phasing Ltd. * * Sheraton House, Castle Park Tel: +44-(0)1223-353033 * * Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 * * * === -- *** Jacob Pearson Keller Northwestern University Medical Scientist Training Program cel: 773.608.9185 email: j-kell...@northwestern.edu ***
Re: [ccp4bb] FW: [ccp4bb] Resolution and distance accuracies
Dear Gerard, Actually, for some of the tests we turned off the DEN network restraints for the last two refinement macrocycles, so, at least for these particular cases, the DEN method truly found a better minimum rather than forcing the system to the higher resolution structure. Cheers, Axel On Dec 23, 2010, at 3:04 PM, Gerard Bricogne wrote: Dear Bernhard, I must say that I find the super-resolution claims in this paper a bit of a conjuring trick. If the final refined model has greater accuracy than one would expect from the resolution of the data it has been refined against, it is because that extra accuracy has been lifted from the higher resolution data that were used to refine the structure on the basis of which the elastic network restraints were created. Should we then say that we achieve super-resolution whenever we refine a macromolecular structure using Engh Huber restraints, because these enable us to achieve distance accuracies comparable with those in the small molecules structures in the Cambridge Structural Database? Perhaps I have missed an essential point of this paper. With best wishes, Gerard. -- On Thu, Dec 23, 2010 at 12:25:26PM -0800, Bernhard Rupp (Hofkristallrat a.D.) wrote: Oops I am outdated: Axel just emailed me that he describes an improved coordinate estimate beyond the Rayleigh criterion in his recent paper Schroder GF, Levitt M, Brunger AT (2010) Super-resolution biomolecular crystallography with low-resolution data. Nature 464(7292), 1218-1222. For the deformable elastic network (DEN) refinement, see his ref 14 Schroder GF, Brunger AT, Levitt M (2007) Combining efficient conformational sampling with a deformable elastic network model facilitates structure refinement at low resolution. Structure 15(12), 1630-1641. BR -- === * * * Gerard Bricogne g...@globalphasing.com * * * * Global Phasing Ltd. * * Sheraton House, Castle Park Tel: +44-(0)1223-353033 * * Cambridge CB3 0AX, UK Fax: +44-(0)1223-366889 * * * ===
