Re: [ccp4bb] change in unit cell volume
On Thu, Mar 07, 2019 at 02:32:53PM -0600, Murpholino Peligro wrote: > Let's say I have a protein crystal from which I collected 30 datasets. If I > plot the unit cell volume per dataset the volume rises. You also have to be sure that (1) there is no significant energy drift during those experiments (which if not modelled could be compensated by a change in refined cell dimensions) (2) each dataset has been collected identically (over the same part of the crystal) and contains enough observations to give a stable refinement of your cell parameters I would also check the values of crystal-detector distance as a function of dose - since these should stay stable and not drift as well. Sometimes instrumentation changes/drifts can be compensated by a change in refined cell dimensions while the physical cell within the crystal actually stays static. Assuming you have high quality data for each of those datasets, a good test is to refine your model to convergence against each of those and see if the final (standard) bonds are systematically shorter or longer than the expected values (e.g. Engh for protein). WhatCheck does a check for that and correlates it with a nice suggestion about wrong cell parameters ... which can highlight e.g. errors in energy or wavelength values as written by beamlines into image headers ;-) Cheers Clemens > My question is: Is there a rule of thumb of some sort* to consider the > initial/final datasets isomorphous still? > > * Something like if the unit cell volume changes more than 1% then the > crystal is not isomorphous. > > My second question is: Meents already said that the unit cell volume > expansion is a consequence of hydrogen gas building up inside the crystal. > But...what if the unit cell volume decreases? Is there an explanation for > that? > > > Thank you very much. > > > > To unsubscribe from the CCP4BB list, click the following link: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 -- *-- * Clemens Vonrhein, Ph.D. vonrhein AT GlobalPhasing DOT com * Global Phasing Ltd., Sheraton House, Castle Park * Cambridge CB3 0AX, UK www.globalphasing.com *-- To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
Re: [ccp4bb] change in unit cell volume
Dear Murpholino Unfortunately no rule of thumb has been established, although I have only seen contraction once and have seen expansion for very many proteins. Seems to depend critically on the particular crystal (and probably its density of stacking imperfections/dislocations) and even crystals of the same protein from the same growth drop do not behave in the same way, as shown in 2 systematic studies of this phenomenon: Ravelli et al., JSR (2002) 9 Murray and Garman, JSR (2002) 9, 347-354. Best wishes Elspeth From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Edward Snell Sent: 07 March 2019 20:56 To: CCP4BB@JISCMAIL.AC.UK Subject: Re: [ccp4bb] change in unit cell volume Hi Murpholino, I’ve looked at this with respect to metals in the protein and found that it was very informative to compare fractional coordinates which compensate for the volume expansion. When that is done, some apparent motions may be simply due to unit cell expansion (waters, metals, ligands etc.), while others can be very specific and produce structural non-isomorphism. I suspect the chemical damage has much more of an impact. Owen, Rudino-Pinera and Garman (PNAS, 2006) recommend an absolute maximum dose of 30 MGy. I’ve not compared cell expansion as a function of dose for a large sample of proteins but in a recent study (conveniently just heading for publication) I have seen a linear ~0.3% expansion per MGy which gives ~1% at 30 MGy. I don’t know if it’s the same for other proteins but I seem to remember a study or two on this and fully expect the authors to give me grief for forgetting them at the moment! Unit cell decreases could possibly be an impact of specific damage to crystallization contacts, off center crystals (if it’s within a data set), detector shifts or energy changes (both unlikely). I’ve not heard of a unit cell decrease being driven by damage but that’s not to say that it doesn’t happen. Best, Eddie PS. Shameless plug –- http://getacrystal.com Edward Snell Ph.D. Biological Small Angle Scattering Theory and Practice, Eaton E. Lattman, Thomas D. Grant, and Edward H. Snell. Available through all good bookshops, or direct from Oxford University Press Director of the NSF BioXFEL Science and Technology Center President and CEO Hauptman-Woodward Medical Research Institute BioInnovations Chaired Professorship, University at Buffalo, SUNY 700 Ellicott Street, Buffalo, NY 14203-1102 hwi.buffalo.edu Phone: (716) 898 8631 Fax: (716) 898 8660 Skype:eddie.snell Email: esn...@hwi.buffalo.edu<mailto:esn...@hwi.buffalo.edu> Webpage: https://hwi.buffalo.edu/scientist-directory/snell/ [cid:image002.png@01D4D58B.E18F91C0] Heisenberg was probably here! From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Murpholino Peligro Sent: Thursday, March 7, 2019 3:33 PM To: CCP4BB@JISCMAIL.AC.UK<mailto:CCP4BB@JISCMAIL.AC.UK> Subject: [ccp4bb] change in unit cell volume Let's say I have a protein crystal from which I collected 30 datasets. If I plot the unit cell volume per dataset the volume rises. My question is: Is there a rule of thumb of some sort* to consider the initial/final datasets isomorphous still? * Something like if the unit cell volume changes more than 1% then the crystal is not isomorphous. My second question is: Meents already said that the unit cell volume expansion is a consequence of hydrogen gas building up inside the crystal. But...what if the unit cell volume decreases? Is there an explanation for that? Thank you very much. To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
Re: [ccp4bb] change in unit cell volume
Hi Murpholino and Eddie, I looked up what we found in our study of the decay of thaumatin crystals at 3 energies (Acta Cryst. (2015). D71, 772-778). Figure 2 shows the unit cell parameter a vs. dose for 5 crystals (all from the same batch). One crystal shows a fairly linear expansion of +0.06% per MGy up to 16 MGy. Two other crystals show about the same initial expansion of 0.06%/MGy up to 8 and 10 MGy which then decreases with one of the two shown no expansion from 13 to 16 MGy. Crystal 4 starts with about 0.04%/MGy up to 2 MGy and shows no expansion from 6 to 16 MGy. Crystal 5 starts expanding at 0.03%/MGy up to 4 MGy, no expansion between 6 and 8 MGy, and contracting between 8 and 15 MGy at a final rate of -0.05%/MGy between 12 and 15 MGy. Fig. 2 shows data measured at 6.33 keV. Similar disparity was observed for the 5 crystals each measured at 12.66 keV and at 19 keV. We were not the first observing contraction instead of expansion. We did not address the question of isomorphism. For our study, the change of unit-cell parameters was only of concern to make sure we used only those reflections that were fully recorded from zero to the maximum dose. I agree with Eddie that the chemical changes due to radiation damage contribute much more to non-isomorphism than the small changes in unit-call parameters. By the way, we found that, on average, at a dose of about 11 MGy the average diffraction intensities had decayed to 70% of the initial value at 12.66 keV and 19 keV and at about 7.5 MGy at 6.33 keV. Much lower than 30 MGy as reported by Owen et al. Others have also reported dose limits similar to ours. Gerd On 07.03.2019 14:56, Edward Snell wrote: Hi Murpholino, I’ve looked at this with respect to metals in the protein and found that it was very informative to compare fractional coordinates which compensate for the volume expansion. When that is done, some apparent motions may be simply due to unit cell expansion (waters, metals, ligands etc.), while others can be very specific and produce structural non-isomorphism. I suspect the chemical damage has much more of an impact. Owen, Rudino-Pinera and Garman (PNAS, 2006) recommend an absolute maximum dose of 30 MGy. I’ve not compared cell expansion as a function of dose for a large sample of proteins but in a recent study (conveniently just heading for publication) I have seen a linear ~0.3% expansion per MGy which gives ~1% at 30 MGy. I don’t know if it’s the same for other proteins but I seem to remember a study or two on this and fully expect the authors to give me grief for forgetting them at the moment! Unit cell decreases could possibly be an impact of specific damage to crystallization contacts, off center crystals (if it’s within a data set), detector shifts or energy changes (both unlikely). I’ve not heard of a unit cell decrease being driven by damage but that’s not to say that it doesn’t happen. Best, Eddie PS. Shameless plug –- http://getacrystal.com Edward Snell Ph.D. Biological Small Angle Scattering Theory and Practice, Eaton E. Lattman, Thomas D. Grant, and Edward H. Snell. Available through all good bookshops, or direct from Oxford University Press Director of the NSF BioXFEL Science and Technology Center President and CEO Hauptman-Woodward Medical Research Institute BioInnovations Chaired Professorship, University at Buffalo, SUNY 700 Ellicott Street, Buffalo, NY 14203-1102 hwi.buffalo.edu Phone: (716) 898 8631 Fax: (716) 898 8660 Skype:eddie.snell Email: esn...@hwi.buffalo.edu<mailto:esn...@hwi.buffalo.edu> Webpage: https://hwi.buffalo.edu/scientist-directory/snell/ [cid:part3.F0916F99.BC7F8DEA@anl.gov] Heisenberg was probably here! From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Murpholino Peligro Sent: Thursday, March 7, 2019 3:33 PM To: CCP4BB@JISCMAIL.AC.UK<mailto:CCP4BB@JISCMAIL.AC.UK> Subject: [ccp4bb] change in unit cell volume Let's say I have a protein crystal from which I collected 30 datasets. If I plot the unit cell volume per dataset the volume rises. My question is: Is there a rule of thumb of some sort* to consider the initial/final datasets isomorphous still? * Something like if the unit cell volume changes more than 1% then the crystal is not isomorphous. My second question is: Meents already said that the unit cell volume expansion is a consequence of hydrogen gas building up inside the crystal. But...what if the unit cell volume decreases? Is there an explanation for that? Thank you very much. To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 To unsu
Re: [ccp4bb] change in unit cell volume
Hi Murpholino, I’ve looked at this with respect to metals in the protein and found that it was very informative to compare fractional coordinates which compensate for the volume expansion. When that is done, some apparent motions may be simply due to unit cell expansion (waters, metals, ligands etc.), while others can be very specific and produce structural non-isomorphism. I suspect the chemical damage has much more of an impact. Owen, Rudino-Pinera and Garman (PNAS, 2006) recommend an absolute maximum dose of 30 MGy. I’ve not compared cell expansion as a function of dose for a large sample of proteins but in a recent study (conveniently just heading for publication) I have seen a linear ~0.3% expansion per MGy which gives ~1% at 30 MGy. I don’t know if it’s the same for other proteins but I seem to remember a study or two on this and fully expect the authors to give me grief for forgetting them at the moment! Unit cell decreases could possibly be an impact of specific damage to crystallization contacts, off center crystals (if it’s within a data set), detector shifts or energy changes (both unlikely). I’ve not heard of a unit cell decrease being driven by damage but that’s not to say that it doesn’t happen. Best, Eddie PS. Shameless plug –- http://getacrystal.com Edward Snell Ph.D. Biological Small Angle Scattering Theory and Practice, Eaton E. Lattman, Thomas D. Grant, and Edward H. Snell. Available through all good bookshops, or direct from Oxford University Press Director of the NSF BioXFEL Science and Technology Center President and CEO Hauptman-Woodward Medical Research Institute BioInnovations Chaired Professorship, University at Buffalo, SUNY 700 Ellicott Street, Buffalo, NY 14203-1102 hwi.buffalo.edu Phone: (716) 898 8631 Fax: (716) 898 8660 Skype:eddie.snell Email: esn...@hwi.buffalo.edu Webpage: https://hwi.buffalo.edu/scientist-directory/snell/ [cid:image001.png@01D4D4FE.49ECE2B0] Heisenberg was probably here! From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Murpholino Peligro Sent: Thursday, March 7, 2019 3:33 PM To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] change in unit cell volume Let's say I have a protein crystal from which I collected 30 datasets. If I plot the unit cell volume per dataset the volume rises. My question is: Is there a rule of thumb of some sort* to consider the initial/final datasets isomorphous still? * Something like if the unit cell volume changes more than 1% then the crystal is not isomorphous. My second question is: Meents already said that the unit cell volume expansion is a consequence of hydrogen gas building up inside the crystal. But...what if the unit cell volume decreases? Is there an explanation for that? Thank you very much. To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1 To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
[ccp4bb] change in unit cell volume
Let's say I have a protein crystal from which I collected 30 datasets. If I plot the unit cell volume per dataset the volume rises. My question is: Is there a rule of thumb of some sort* to consider the initial/final datasets isomorphous still? * Something like if the unit cell volume changes more than 1% then the crystal is not isomorphous. My second question is: Meents already said that the unit cell volume expansion is a consequence of hydrogen gas building up inside the crystal. But...what if the unit cell volume decreases? Is there an explanation for that? Thank you very much. To unsubscribe from the CCP4BB list, click the following link: https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
