Re: [ccp4bb] Do my SAXS data agree with the crystal structure?
Dear Xun, Regarding your monomer vs dimer, theoretical vs observed crysol plots - yes - they are significantly different. If you focus at the very lowest q part of the curve - the deviation there in your monomer plots indicate that there is a significant size difference between your PX monomer and your SAXS data - the PX dimer is a much better fit at low q. This should be enough to demonstrate to a reviewer that the dimer you see in PX is also present in solution. Other experiments that could support this are SEC-MALLS or perhaps AUC. HTH, Dave David C. Briggs PhD Father, Structural Biologist and Sceptic University of Manchester E-mail: david.c.bri...@manchester.ac.uk Webs : http://flavors.me/xtaldave Twitter: @xtaldave Skype: DocDCB On 17 June 2012 06:11, Xun Lu xlun...@gmail.com wrote: Drs.Caldwell, Briggs, and Gupta, Thank you very much for the advices. I regret that I didn't show any figure in the earlier post. Here I've attached a figure showing the data quality and some fittings. Data look OK, right? This question may sound silly, but I just want to make sure. As I said in the earlier post, I tried Crysol. I used the crystal structure (dimer+DNA) as the model, and the fitting was OK, right? In fact, I also tried monomer+DNA as the model (I simply deleted one monomer from the PDB file). This kind of comparison may be meaningless, but I was just curious. I am wondering how people judge whether the fit is good or not. Another question, I tried to generate an envelope from SAXS data using Gasbor and Dammin (people say Dammin is better at protein-DNA complex, although it still uses the same bead for both DNA and protein?). The generated envelope was nothing like my crystal structure. As people have pointed out, protein and DNA scatter differently. SANS is the way to go. So I should give up on modeling SAXS data? I've almost given up, because anyways I have the crystal structure, and SAXS is only a small part of this paper. Thanks, Xun On Sat, Jun 16, 2012 at 6:36 PM, Kushol Gupta kushol.gu...@gmail.com wrote: Two cents - A good deal of caution must be exercised when working with composite particles such as a protein-DNA complex in SAXS because of the contrast problem. Simply, protein and DNA scatter differently in x-rays, with a bias towards the DNA component. As a result, experimental Rgs could be slightly deflated versus what their true values would be at infinite contrast. Mass estimation by I(0) analysis with a protein standard of known mass and concentration is not really valid because the contrast terms are different. Because the particle is heterogeneous in composition and distribution, shape reconstruction from SAXS alone, which assumes homogeneity, can also be misleading (although in practice it is still reasonably instructive). It is for these reasons that SANS and the contrast variation approach can be extremely useful. With those caveats, the strategy you describe - comparison of experimental and theoretical profiles from an experimental structure using CRYSOL or FoxS is definitely the best way to go in the case of a protein-DNA complex with SAXS alone. Showing comparisons of the experimental with the calculated should make the point. Test other possible models inferred from lattice packing to further your point (if applicable). Regarding populations of monomer and dimer - · it is generally good to constrain your interpretation of scattering data with other orthogonal solution measures which demonstrates the homogeneity of your complex in comparable experimental conditions, such as sedimentation velocity or gel filtration. · Have some determination of affinity of the complex in the same solution conditions (including temperature!). This will allow you to argue that your sample concentrations are well in excess of any monomer-dimer association behavior (eg, mixtures!). Scattering of mixtures can undermine your ability to accurately assess the structural properties of your complex. · Collect a concentration series and extrapolate to infinite dilution, if possible, to ensure elimination of the S(q) term from your data. Interparticle interactions can be an issue with complexes containing DNA if the buffers aren’t quite right. (I’ve seen this a lot) Lastly, remember that the scattering profile represents the solution average of the particle, not just a single snapshot. Some discrepancies like those you note should be expected. Hope that helps, Kushol Kushol Gupta, Ph.D. Research Associate - Van Duyne Laboratory HHMI / Perelman School of Medicine University of Pennsylvania kgu...@mail.med.upenn.edu 215-573-7260 / 267-259-0082 -Original Message- From: CCP4 bulletin board
Re: [ccp4bb] Do my SAXS data agree with the crystal structure?
If you main point is dimer vs monomer, Guinier plots is in fact all you would need, in my opinion according to what I have read so far... A. On 17 Jun 2012, at 13:01, David Briggs wrote: Dear Xun, Regarding your monomer vs dimer, theoretical vs observed crysol plots - yes - they are significantly different. If you focus at the very lowest q part of the curve - the deviation there in your monomer plots indicate that there is a significant size difference between your PX monomer and your SAXS data - the PX dimer is a much better fit at low q. This should be enough to demonstrate to a reviewer that the dimer you see in PX is also present in solution. Other experiments that could support this are SEC-MALLS or perhaps AUC. HTH, Dave David C. Briggs PhD Father, Structural Biologist and Sceptic University of Manchester E-mail: david.c.bri...@manchester.ac.uk Webs : http://flavors.me/xtaldave Twitter: @xtaldave Skype: DocDCB On 17 June 2012 06:11, Xun Lu xlun...@gmail.com wrote: Drs.Caldwell, Briggs, and Gupta, Thank you very much for the advices. I regret that I didn't show any figure in the earlier post. Here I've attached a figure showing the data quality and some fittings. Data look OK, right? This question may sound silly, but I just want to make sure. As I said in the earlier post, I tried Crysol. I used the crystal structure (dimer+DNA) as the model, and the fitting was OK, right? In fact, I also tried monomer+DNA as the model (I simply deleted one monomer from the PDB file). This kind of comparison may be meaningless, but I was just curious. I am wondering how people judge whether the fit is good or not. Another question, I tried to generate an envelope from SAXS data using Gasbor and Dammin (people say Dammin is better at protein-DNA complex, although it still uses the same bead for both DNA and protein?). The generated envelope was nothing like my crystal structure. As people have pointed out, protein and DNA scatter differently. SANS is the way to go. So I should give up on modeling SAXS data? I've almost given up, because anyways I have the crystal structure, and SAXS is only a small part of this paper. Thanks, Xun On Sat, Jun 16, 2012 at 6:36 PM, Kushol Gupta kushol.gu...@gmail.com wrote: Two cents - A good deal of caution must be exercised when working with composite particles such as a protein-DNA complex in SAXS because of the contrast problem. Simply, protein and DNA scatter differently in x-rays, with a bias towards the DNA component. As a result, experimental Rgs could be slightly deflated versus what their true values would be at infinite contrast. Mass estimation by I(0) analysis with a protein standard of known mass and concentration is not really valid because the contrast terms are different. Because the particle is heterogeneous in composition and distribution, shape reconstruction from SAXS alone, which assumes homogeneity, can also be misleading (although in practice it is still reasonably instructive). It is for these reasons that SANS and the contrast variation approach can be extremely useful. With those caveats, the strategy you describe - comparison of experimental and theoretical profiles from an experimental structure using CRYSOL or FoxS is definitely the best way to go in the case of a protein-DNA complex with SAXS alone. Showing comparisons of the experimental with the calculated should make the point. Test other possible models inferred from lattice packing to further your point (if applicable). Regarding populations of monomer and dimer - · it is generally good to constrain your interpretation of scattering data with other orthogonal solution measures which demonstrates the homogeneity of your complex in comparable experimental conditions, such as sedimentation velocity or gel filtration. · Have some determination of affinity of the complex in the same solution conditions (including temperature!). This will allow you to argue that your sample concentrations are well in excess of any monomer-dimer association behavior (eg, mixtures!). Scattering of mixtures can undermine your ability to accurately assess the structural properties of your complex. · Collect a concentration series and extrapolate to infinite dilution, if possible, to ensure elimination of the S(q) term from your data. Interparticle interactions can be an issue with complexes containing DNA if the buffers aren’t quite right. (I’ve seen this a lot) Lastly, remember that the scattering profile represents the solution average of the particle, not just a single snapshot. Some discrepancies like those you note should be expected. Hope that helps, Kushol Kushol
Re: [ccp4bb] Do my SAXS data agree with the crystal structure?
It might also be useful to go through the recent paper by Jacques et al Acta cryst D68:620-6 (2012) towards the standardization of saxs data analysis, diagnostics and reporting. Best regards, Savvas On 17-jun.-2012, at 13:01, David Briggs drdavidcbri...@gmail.com wrote: Dear Xun, Regarding your monomer vs dimer, theoretical vs observed crysol plots - yes - they are significantly different. If you focus at the very lowest q part of the curve - the deviation there in your monomer plots indicate that there is a significant size difference between your PX monomer and your SAXS data - the PX dimer is a much better fit at low q. This should be enough to demonstrate to a reviewer that the dimer you see in PX is also present in solution. Other experiments that could support this are SEC-MALLS or perhaps AUC. HTH, Dave David C. Briggs PhD Father, Structural Biologist and Sceptic University of Manchester E-mail: david.c.bri...@manchester.ac.uk Webs : http://flavors.me/xtaldave Twitter: @xtaldave Skype: DocDCB On 17 June 2012 06:11, Xun Lu xlun...@gmail.com wrote: Drs.Caldwell, Briggs, and Gupta, Thank you very much for the advices. I regret that I didn't show any figure in the earlier post. Here I've attached a figure showing the data quality and some fittings. Data look OK, right? This question may sound silly, but I just want to make sure. As I said in the earlier post, I tried Crysol. I used the crystal structure (dimer+DNA) as the model, and the fitting was OK, right? In fact, I also tried monomer+DNA as the model (I simply deleted one monomer from the PDB file). This kind of comparison may be meaningless, but I was just curious. I am wondering how people judge whether the fit is good or not. Another question, I tried to generate an envelope from SAXS data using Gasbor and Dammin (people say Dammin is better at protein-DNA complex, although it still uses the same bead for both DNA and protein?). The generated envelope was nothing like my crystal structure. As people have pointed out, protein and DNA scatter differently. SANS is the way to go. So I should give up on modeling SAXS data? I've almost given up, because anyways I have the crystal structure, and SAXS is only a small part of this paper. Thanks, Xun On Sat, Jun 16, 2012 at 6:36 PM, Kushol Gupta kushol.gu...@gmail.com wrote: Two cents - A good deal of caution must be exercised when working with composite particles such as a protein-DNA complex in SAXS because of the contrast problem. Simply, protein and DNA scatter differently in x-rays, with a bias towards the DNA component. As a result, experimental Rgs could be slightly deflated versus what their true values would be at infinite contrast. Mass estimation by I(0) analysis with a protein standard of known mass and concentration is not really valid because the contrast terms are different. Because the particle is heterogeneous in composition and distribution, shape reconstruction from SAXS alone, which assumes homogeneity, can also be misleading (although in practice it is still reasonably instructive). It is for these reasons that SANS and the contrast variation approach can be extremely useful. With those caveats, the strategy you describe - comparison of experimental and theoretical profiles from an experimental structure using CRYSOL or FoxS is definitely the best way to go in the case of a protein-DNA complex with SAXS alone. Showing comparisons of the experimental with the calculated should make the point. Test other possible models inferred from lattice packing to further your point (if applicable). Regarding populations of monomer and dimer - · it is generally good to constrain your interpretation of scattering data with other orthogonal solution measures which demonstrates the homogeneity of your complex in comparable experimental conditions, such as sedimentation velocity or gel filtration. · Have some determination of affinity of the complex in the same solution conditions (including temperature!). This will allow you to argue that your sample concentrations are well in excess of any monomer-dimer association behavior (eg, mixtures!). Scattering of mixtures can undermine your ability to accurately assess the structural properties of your complex. · Collect a concentration series and extrapolate to infinite dilution, if possible, to ensure elimination of the S(q) term from your data. Interparticle interactions can be an issue with complexes containing DNA if the buffers aren’t quite right. (I’ve seen this a lot) Lastly, remember that the scattering profile represents the solution average of the particle, not just a single snapshot. Some discrepancies like
Re: [ccp4bb] Do my SAXS data agree with the crystal structure?
Hi Xun, it is difficult to judge without seeing the P(r) plots, but seeing as you have a dimer in your crystal structure and a dimer in your SAXS, AND your Chi2 value seems reasonable for a good match between PX and SAXS, I'd say you've got what you need. A matching P(r) plot would be nice, but the crysol plot comparing your SAXS scattering curve and the theoretical scattering curve from your crystal structure should be enough for a figure. HTH, Dave David C. Briggs PhD Father, Structural Biologist and Sceptic University of Manchester E-mail: david.c.bri...@manchester.ac.uk Webs : http://flavors.me/xtaldave Twitter: @xtaldave Skype: DocDCB On 16 June 2012 19:29, Xun Lu xlun...@gmail.com wrote: Dear all, I have solved a protein-DNA structure, and I also did SAXS to get some ideas of the solution structure. The SAXS data were good, no aggregation at all three tested concentrations. I tried to use Crysol to see if my crystal structure fits the SAXS. The fitting to the scattering profile seems good to me and the Chi2 is 1~1.4. Then I wanted to see how the P(r) looked like (wanted to make a figure for my paper:). I calculated the theoretical scattering profile of the crystal structure from an online server (FOXS). I then run GNOM to make P(r). To my surprise, this theoretical P(r) looks a little different from the P(r) of SAXS data. There's a very small bump that was peaked at 70A (Dmax is 108A, which seems reasonable from the crystal structure). The major peak was at 25A. As some people said, P(r) is indeed quite sensitive to subtle differences. The protein is a dimer in the crystal, although it can also bind DNA as a monomer (much more loosely). The estimated MW from SAXS indicates it's a dimer in solution as well. It seems that I got the information I wanted from the SAXS experiment, but maybe not. Due to the low resolution of SAXS, maybe I can only say that the majority is a dimer?? Would it be possible to see the monomer if there's only 10% of them in the solution? How to interpret the discrepancy between the P(r) from crystal and the P(r) from SAXS? Any comments are welcome! Xun Sent from my iPad
Re: [ccp4bb] Do my SAXS data agree with the crystal structure?
Two cents - A good deal of caution must be exercised when working with composite particles such as a protein-DNA complex in SAXS because of the contrast problem. Simply, protein and DNA scatter differently in x-rays, with a bias towards the DNA component. As a result, experimental Rgs could be slightly deflated versus what their true values would be at infinite contrast. Mass estimation by I(0) analysis with a protein standard of known mass and concentration is not really valid because the contrast terms are different. Because the particle is heterogeneous in composition and distribution, shape reconstruction from SAXS alone, which assumes homogeneity, can also be misleading (although in practice it is still reasonably instructive). It is for these reasons that SANS and the contrast variation approach can be extremely useful. With those caveats, the strategy you describe - comparison of experimental and theoretical profiles from an experimental structure using CRYSOL or FoxS is definitely the best way to go in the case of a protein-DNA complex with SAXS alone. Showing comparisons of the experimental with the calculated should make the point. Test other possible models inferred from lattice packing to further your point (if applicable). Regarding populations of monomer and dimer - . it is generally good to constrain your interpretation of scattering data with other orthogonal solution measures which demonstrates the homogeneity of your complex in comparable experimental conditions, such as sedimentation velocity or gel filtration. . Have some determination of affinity of the complex in the same solution conditions (including temperature!). This will allow you to argue that your sample concentrations are well in excess of any monomer-dimer association behavior (eg, mixtures!). Scattering of mixtures can undermine your ability to accurately assess the structural properties of your complex. . Collect a concentration series and extrapolate to infinite dilution, if possible, to ensure elimination of the S(q) term from your data. Interparticle interactions can be an issue with complexes containing DNA if the buffers aren't quite right. (I've seen this a lot) Lastly, remember that the scattering profile represents the solution average of the particle, not just a single snapshot. Some discrepancies like those you note should be expected. Hope that helps, Kushol Kushol Gupta, Ph.D. Research Associate - Van Duyne Laboratory HHMI / Perelman School of Medicine University of Pennsylvania kgu...@mail.med.upenn.edu 215-573-7260 / 267-259-0082 -Original Message- From: CCP4 bulletin board [mailto:CCP4BB@JISCMAIL.AC.UK] On Behalf Of Xun Lu Sent: Saturday, June 16, 2012 2:29 PM To: CCP4BB@JISCMAIL.AC.UK Subject: [ccp4bb] Do my SAXS data agree with the crystal structure? Dear all, I have solved a protein-DNA structure, and I also did SAXS to get some ideas of the solution structure. The SAXS data were good, no aggregation at all three tested concentrations. I tried to use Crysol to see if my crystal structure fits the SAXS. The fitting to the scattering profile seems good to me and the Chi2 is 1~1.4. Then I wanted to see how the P(r) looked like (wanted to make a figure for my paper:). I calculated the theoretical scattering profile of the crystal structure from an online server (FOXS). I then run GNOM to make P(r). To my surprise, this theoretical P(r) looks a little different from the P(r) of SAXS data. There's a very small bump that was peaked at 70A (Dmax is 108A, which seems reasonable from the crystal structure). The major peak was at 25A. As some people said, P(r) is indeed quite sensitive to subtle differences. The protein is a dimer in the crystal, although it can also bind DNA as a monomer (much more loosely). The estimated MW from SAXS indicates it's a dimer in solution as well. It seems that I got the information I wanted from the SAXS experiment, but maybe not. Due to the low resolution of SAXS, maybe I can only say that the majority is a dimer?? Would it be possible to see the monomer if there's only 10% of them in the solution? How to interpret the discrepancy between the P(r) from crystal and the P(r) from SAXS? Any comments are welcome! Xun Sent from my iPad=
