Hi guys, I'm a bit confused by this answer. I get the "add dummy atoms and calculate map" to check whether it is Fourier truncation ripples (which I don't think it will turn out to be). But I wouldn't feel comfortable depositing a structure with dummy atoms even if they do have zero occupancy. Are you really suggesting that people do that? Secondly, when I look in the .def for my refinements I find two entries for mask calculation: Under the fake_f_obs heading mask { solvent_radius = 1.11 shrink_truncation_radius = 0.9 grid_step_factor = 4 verbose = 1 mean_shift_for_mask_update = 0.1 ignore_zero_occupancy_atoms = True ignore_hydrogens = True } And again under it's own heading towards the end mask { solvent_radius = 1.11 shrink_truncation_radius = 0.9 grid_step_factor = 4 verbose = 1 mean_shift_for_mask_update = 0.1 ignore_zero_occupancy_atoms = True ignore_hydrogens = True } Which one is relevant? Also why didn't any of you suggest the optimize_mask=true parameter? Shouldn't that automatically find the best solvent_radius and shrink_truncation_radius values? Sorry if these are dumb questions (and sorry that there are so many) but I was just really confused by these answers. Sincerely, Morten Grøftehauge 2008/10/4 Pavel Afonine

Hi Frank,

I just want to add to Ralf's very comprehensive reply... The parameters solvent_radius, shrink_truncation_radius and grid_step_factor are explained in the original paper:

Jiang, J.-S. & Brünger, A. T. (1994). J. Mol. Biol. 243, 100-115. "Protein hydration observed by X-ray diffraction. Solvation properties of penicillopepsin and neuraminidase crystal structures."

The details of PHENIX implementation of this are described here:

P.V. Afonine, R.W. Grosse-Kunstleve & P.D. Adams. Acta Cryst. (2005). D61, 850-855. "A robust bulk-solvent correction and anisotropic scaling procedure"

Also, the negative peaks you observe can easily be Fourier series truncation ripples. I think Ralf's suggestion to place some dummy atoms there with zero occupancy is a good idea. I wouldn't even do any refinement (since moving atoms may cancel these artifacts), but just compute two maps - with and w/o the dummy atoms and see what happens to these negative peaks.

Cheers, Pavel.

On 9/28/2008 3:25 PM, Frank von Delft wrote:

...

Hi

After being through phenix.refine, I see in my hydrophobic core a big space (a few atoms wide) that is filled with strong negative difference density. I suspect the culprit is the bulk solvent mask, which is defined too tightly.

The online manual mentions three parameters, but not what they do. solvent_radius, shrink_truncation_radius, grid_step_factor

What *exactly* do they do?

(I thought I'd elicit a contribution for the online docs this way :) Cheers phx _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

-- Morten K Grøftehauge PhD student Department of Molecular Biology Gustav Wieds Vej 10 C 8000 Aarhus C - Denmark Phone: +45 89 42 52 61 Fax: +45 86 12 31 78 www.bioxray.dk

I know this has been asked before, and there are ways around to get this done in Autosol or Xtriage, but a sharpening B-factor for low resolution maps would be a very nice addition to phenix.refine. Are there any plans to add a parameter in phenix.refine for this? Best, Engin

Not personally. Two people in the lab have done it and are significantly benefiting (and I cannot volunteer their datasets, since the structures are not finished or published yet). However, you can probably get from the pdb database the Delabarre and Brunger data for p97/VCP at 4.4 A (PDB ID: 1YQ0) as explained in Acta Cryst D 62 p. 923. It is implemented in CNS following this paper. Among our lab people the two cases where it really worked were both 3.5 A datasets with experimental phasing (as the Brunger papers also point out to), and in cases without experimental phasing, I have not seen much improvement. The magical values in one case was -100 A^2, and in the other -50 to -100 (For the published Brunger data, it was -120). Thank you, Engin Pavel Afonine wrote:

Hi Engin,

yes, it is in the list of things to do.

Do you have a specific example of where B-factor sharpening really helps (data, model, and the magic value for B-sharp) that I can use as a reference?

Pavel.

On 1/23/2009 1:43 PM, Engin Ozkan wrote:

...

I know this has been asked before, and there are ways around to get this done in Autosol or Xtriage, but a sharpening B-factor for low resolution maps would be a very nice addition to phenix.refine. Are there any plans to add a parameter in phenix.refine for this?

Best,

Engin _______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

Ok, just to wrap this up. I added zero occupancy dummy atoms to the structure and the R-free improved by about 0.5 and more importantly the Fo-Fc looked better in other areas of the map. So I am including the dummy atoms for refinement and then of course removing them before the final refinement. I don't know what Frank Von Delft chose to do. I addition it seems that Brian W. Matthews and Lijun Liu have review coming out in Protein Science about this subject: "A review about nothing: Are apolar cavities in proteins really empty?" I haven't read it yet but it seems that the answer is that no-one knows. Cheers, Morten 2009/1/6 Pavel Afonine

Hi Morten,

...

I get the "add dummy atoms and calculate map" to check whether it is Fourier truncation ripples (which I don't think it will turn out to be). But I wouldn't feel comfortable depositing a structure with dummy atoms even if they do have zero occupancy. Are you really suggesting that people do that?

Of course not. The above suggestion was to test the idea about where the negative peaks are coming from (just to mask the region in order to avoid placing a flat bulk solvent density there and see if the peaks are still there).

Just for your information: there is a number of files in PDB that contain "dummy atoms" which is obviously bad (since scattering type is undefined which makes it impossible to compute the R-factors for such models). Here are a few examples: 2aed, 2ajp, 2amx, 2ar0, 1au9, 2awp, 2ayv, 2b25, ... I can go on, I have a complete list...

...

Secondly, when I look in the .def for my refinements I find two entries for mask calculation: Under the fake_f_obs heading mask { solvent_radius = 1.11 shrink_truncation_radius = 0.9 grid_step_factor = 4 verbose = 1 mean_shift_for_mask_update = 0.1 ignore_zero_occupancy_atoms = True ignore_hydrogens = True } And again under it's own heading towards the end mask { solvent_radius = 1.11 shrink_truncation_radius = 0.9 grid_step_factor = 4 verbose = 1 mean_shift_for_mask_update = 0.1 ignore_zero_occupancy_atoms = True ignore_hydrogens = True }

Which one is relevant?

Both are relevant, but used for different purposes. The first one is used for experimenting/development/testing ideas, when real Fobs are replaced with calculated ones (as the scope name suggests: fake_f_obs): fake_Fobs = Fmodel = scale_k1 * exp(-h*U_overall*ht) * (Fcalc + k_sol * exp(-B_sol*s^2) * Fmask)

The second scope defines the mask parameters for everything else and this is the one that you most likely want to play with.

...

Also why didn't any of you suggest the optimize_mask=true parameter? Shouldn't that automatically find the best solvent_radius and shrink_truncation_radius values?

It's a good idea to try out, but that mask optimization does a grid search over a range of solvent_radius and shrink_truncation_radius and the "best" values are chosen based on the lowest R (or Rfree, I don't remember exactly). I wouldn't believe that fixing a few negative peaks somewhere will show up in R-factor (R-factor is a global measure and very insensitive to minor local changes).

Cheers, Pavel.

_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

-- Morten K Grøftehauge PhD student Department of Molecular Biology Gustav Wieds Vej 10 C 8000 Aarhus C - Denmark Phone: +45 89 42 52 61 Fax: +45 86 12 31 78 www.bioxray.dk

Hi Morten, I think in your case it is very likely (but not 100% sure) that those peaks were due to some bulk solvent mask artifacts. Remember, when we add dummy atoms there with zero occupancy (which means they do not contribute to the scattering at all) (the only change they produced was the bulk solvent mask), these peaks disappears. Pavel. On 1/20/2009 3:47 AM, Morten Grøftehauge wrote:

Ok, just to wrap this up.

I added zero occupancy dummy atoms to the structure and the R-free improved by about 0.5 and more importantly the Fo-Fc looked better in other areas of the map. So I am including the dummy atoms for refinement and then of course removing them before the final refinement. I don't know what Frank Von Delft chose to do.

I addition it seems that Brian W. Matthews and Lijun Liu have review coming out in Protein Science about this subject: "A review about nothing: Are apolar cavities in proteins really empty?" I haven't read it yet but it seems that the answer is that no-one knows.

Cheers, Morten

2009/1/6 Pavel Afonine mailto:[email protected]>

Hi Morten,

> I get the "add dummy atoms and calculate map" to check whether it is > Fourier truncation ripples (which I don't think it will turn out to be). > But I wouldn't feel comfortable depositing a structure with dummy > atoms even if they do have zero occupancy. Are you really suggesting > that people do that?

Of course not. The above suggestion was to test the idea about where the negative peaks are coming from (just to mask the region in order to avoid placing a flat bulk solvent density there and see if the peaks are still there).

Just for your information: there is a number of files in PDB that contain "dummy atoms" which is obviously bad (since scattering type is undefined which makes it impossible to compute the R-factors for such models). Here are a few examples: 2aed, 2ajp, 2amx, 2ar0, 1au9, 2awp, 2ayv, 2b25, ... I can go on, I have a complete list...

> Secondly, when I look in the .def for my refinements I find two > entries for mask calculation: > Under the fake_f_obs heading > mask { > solvent_radius = 1.11 > shrink_truncation_radius = 0.9 > grid_step_factor = 4 > verbose = 1 > mean_shift_for_mask_update = 0.1 > ignore_zero_occupancy_atoms = True > ignore_hydrogens = True > } > And again under it's own heading towards the end > mask { > solvent_radius = 1.11 > shrink_truncation_radius = 0.9 > grid_step_factor = 4 > verbose = 1 > mean_shift_for_mask_update = 0.1 > ignore_zero_occupancy_atoms = True > ignore_hydrogens = True > } > > Which one is relevant?

Both are relevant, but used for different purposes. The first one is used for experimenting/development/testing ideas, when real Fobs are replaced with calculated ones (as the scope name suggests: fake_f_obs): fake_Fobs = Fmodel = scale_k1 * exp(-h*U_overall*ht) * (Fcalc + k_sol * exp(-B_sol*s^2) * Fmask)

The second scope defines the mask parameters for everything else and this is the one that you most likely want to play with.

> Also why didn't any of you suggest the optimize_mask=true parameter? > Shouldn't that automatically find the best solvent_radius and > shrink_truncation_radius values?

It's a good idea to try out, but that mask optimization does a grid search over a range of solvent_radius and shrink_truncation_radius and the "best" values are chosen based on the lowest R (or Rfree, I don't remember exactly). I wouldn't believe that fixing a few negative peaks somewhere will show up in R-factor (R-factor is a global measure and very insensitive to minor local changes).

Cheers, Pavel.

_______________________________________________ phenixbb mailing list [email protected] mailto:[email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb

-- Morten K Grøftehauge PhD student Department of Molecular Biology Gustav Wieds Vej 10 C 8000 Aarhus C - Denmark Phone: +45 89 42 52 61 Fax: +45 86 12 31 78 www.bioxray.dk http://www.bioxray.dk

------------------------------------------------------------------------

_______________________________________________ phenixbb mailing list [email protected] http://www.phenix-online.org/mailman/listinfo/phenixbb