Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Jose Brandao-Neto]

Dear Samuel,

 I'm guessing you are not constrained by sample availability. Consider a 
multi-crystal approach to understand your native structure and contrast against 
structures with low occupancy binders (Pearce et al 2017, 
https://www.nature.com/articles/ncomms15123 ). 

 That structure (a pandda ground state model) will be a great place to start a 
polder map.

Jose'

===
Jose Brandao-Neto MPhil CPhys
Senior Beamline Scientist - I04-1
XChem Scientist
Diamond Light Source

jose.brandao-n...@diamond.ac.uk
+44 (0)1235 778506
===



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Pavel Afonine]

Hi,
this is why Polder map tool also includes analysis of the map in question
to determine whether it looks like bulk-solvent or something else, as
described in paragraph 5 here:
http://journals.iucr.org/d/issues/2017/02/00/ba5254/ba5254.pdf
This analysis tells you in plain English what you are likely to look at.
Otherwise I agree that making sense of maps in solvent-excluded regions are
very tricky.
Pavel

On Fri, Feb 8, 2019 at 2:32 AM Bernhard Rupp 
wrote:

> Hi Fellows,
>
> I'd really like to emphasize the point in the Buster instructions "be
> careful when
> examining fo-fc at low levels" when solvent is excluded. If the solvent
> contribution
> is omitted where you suspect the ligand (e.g. occupancy 0.02 in Refmac),
> there
> will be a fo contribution there from the solvent that is there.
> Particularly if that solvent is a
> dense solution, that fo component will show up nicely and at not so low
> difference
> map levels, and in the shape of that 'excluded' ligand.
>
> Figure 2 in link illustrates that.
> https://febs.onlinelibrary.wiley.com/doi/epdf/10./febs.14320
>
> If you start to fill that void with multiple ligands of various low
> occupancies, you
> are effectively modelling disordered solvent. This is particularly tempting
> because
> I found multiple cases where the classical RSR and RSCC measures give
> acceptable stats for such models. The hunt for low occupancy ligands then
> quickly
> becomes murky density fishing business...
>
> Best, BR
>
>
> -Original Message-
> From: CCP4 bulletin board  On Behalf Of Clemens
> Vonrhein
> Sent: Friday, February 8, 2019 09:53
> To: CCP4BB@JISCMAIL.AC.UK
> Subject: Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to
> generate Polder maps?
>
> Dear Samuel,
>
> On Mon, Feb 04, 2019 at 11:39:58AM +, Samuel Davis (PG Research) wrote:
> > I'm wondering if anyone knows if it is possible to turn off the bulk
> > solvent modelling in Refmac5, for the purpose of generating Polder
> > maps? I know that an option for Polder maps is directly implemented in
> > Phenix, but we ideally want to use Refmac5, as we have used it for the
> > rest of our refinement and want to keep it consistent if possible.
>
> And if you want to try the original implementation of the underlying idea
> as
> an alternative, have a look at the ligand detection mode and maps [1]
> produced by BUSTER [2]. See also [3] and some early examples of their
> usefulness [4-5].
>
> Cheers
>
> Clemens
>
> [1]
> https://www.globalphasing.com/buster/wiki/index.cgi?LigandDetectionModes
> [2] https://www.globalphasing.com/buster/
> [3] Vonrhein, C., & Bricogne, G. (2005). "Automated Structure
> Refinement for High-throughput Ligand Detection with
> BUSTER-TNT". Acta Crysta A61, C248.
> [4] Thoma, Ralf, et al. "Insight into steroid scaffold formation from
> the structure of human oxidosqualene cyclase." Nature 432.7013
> (2004): 118.
> [5] Ekroos, Marika, and Tove Sjogren. "Structural basis for ligand
> promiscuity in cytochrome P450 3A4." Proceedings of the National
> Academy of Sciences 103.37 (2006): 13682-13687.
>
> --
>
> *--
> * 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
>
> 
>
> 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] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Bernhard Rupp]

Hi Fellows,

I'd really like to emphasize the point in the Buster instructions "be
careful when
examining fo-fc at low levels" when solvent is excluded. If the solvent
contribution
is omitted where you suspect the ligand (e.g. occupancy 0.02 in Refmac),
there 
will be a fo contribution there from the solvent that is there.
Particularly if that solvent is a
dense solution, that fo component will show up nicely and at not so low
difference 
map levels, and in the shape of that 'excluded' ligand. 

Figure 2 in link illustrates that. 
https://febs.onlinelibrary.wiley.com/doi/epdf/10./febs.14320

If you start to fill that void with multiple ligands of various low
occupancies, you
are effectively modelling disordered solvent. This is particularly tempting
because 
I found multiple cases where the classical RSR and RSCC measures give 
acceptable stats for such models. The hunt for low occupancy ligands then
quickly 
becomes murky density fishing business...
  
Best, BR


-Original Message-
From: CCP4 bulletin board  On Behalf Of Clemens
Vonrhein
Sent: Friday, February 8, 2019 09:53
To: CCP4BB@JISCMAIL.AC.UK
Subject: Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to
generate Polder maps?

Dear Samuel,

On Mon, Feb 04, 2019 at 11:39:58AM +, Samuel Davis (PG Research) wrote:
> I'm wondering if anyone knows if it is possible to turn off the bulk 
> solvent modelling in Refmac5, for the purpose of generating Polder 
> maps? I know that an option for Polder maps is directly implemented in 
> Phenix, but we ideally want to use Refmac5, as we have used it for the 
> rest of our refinement and want to keep it consistent if possible.

And if you want to try the original implementation of the underlying idea as
an alternative, have a look at the ligand detection mode and maps [1]
produced by BUSTER [2]. See also [3] and some early examples of their
usefulness [4-5].

Cheers

Clemens

[1] https://www.globalphasing.com/buster/wiki/index.cgi?LigandDetectionModes
[2] https://www.globalphasing.com/buster/
[3] Vonrhein, C., & Bricogne, G. (2005). "Automated Structure
Refinement for High-throughput Ligand Detection with
BUSTER-TNT". Acta Crysta A61, C248.
[4] Thoma, Ralf, et al. "Insight into steroid scaffold formation from
the structure of human oxidosqualene cyclase." Nature 432.7013
(2004): 118.
[5] Ekroos, Marika, and Tove Sjogren. "Structural basis for ligand
promiscuity in cytochrome P450 3A4." Proceedings of the National
Academy of Sciences 103.37 (2006): 13682-13687.

-- 

*--
* 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



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Clemens Vonrhein]

Dear Samuel,

On Mon, Feb 04, 2019 at 11:39:58AM +, Samuel Davis (PG Research) wrote:
> I'm wondering if anyone knows if it is possible to turn off the bulk
> solvent modelling in Refmac5, for the purpose of generating Polder
> maps? I know that an option for Polder maps is directly implemented
> in Phenix, but we ideally want to use Refmac5, as we have used it
> for the rest of our refinement and want to keep it consistent if
> possible.

And if you want to try the original implementation of the underlying
idea as an alternative, have a look at the ligand detection mode and
maps [1] produced by BUSTER [2]. See also [3] and some early examples
of their usefulness [4-5].

Cheers

Clemens

[1] https://www.globalphasing.com/buster/wiki/index.cgi?LigandDetectionModes
[2] https://www.globalphasing.com/buster/
[3] Vonrhein, C., & Bricogne, G. (2005). "Automated Structure
Refinement for High-throughput Ligand Detection with
BUSTER-TNT". Acta Crysta A61, C248.
[4] Thoma, Ralf, et al. "Insight into steroid scaffold formation from
the structure of human oxidosqualene cyclase." Nature 432.7013
(2004): 118.
[5] Ekroos, Marika, and Tove Sjogren. "Structural basis for ligand
promiscuity in cytochrome P450 3A4." Proceedings of the National
Academy of Sciences 103.37 (2006): 13682-13687.

-- 

*--
* 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] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Holton, James M]

Yes, Dorothee is right.  Don't turn off all the bulk solvent!  That is not a 
polder map.

In refmac you want to use the keyword:
solvent exclude DUM

And then fill the space you want to have no bulk solvent with water atoms with 
residue ID "DUM" and perhaps occupancy set to zero, since you don't want them 
contributing to Fcalc.  You might also want to turn off clashes for these dummy 
atoms using:
vdwrestraints DUMM 0
or
vdwrestraints exlcude

documentation is here:
https://www2.mrc-lmb.cam.ac.uk/groups/murshudov/content/refmac/refmac_keywords.html

Refmac also supports externally-defined solvent, which can be very handy.  All 
you need is structure factors (amplitudes and phases) in the same MTZ file as 
your data.  I like to call them Fsolvent PHIsolvent. Put these on the LABIN 
line with FPART1=Fsolvent PHIP1=PHIsolvent.  You will probably also want to use:
SCPART 1
Tells refmac to optimize the scale and B factor assigned to your bulk solvent.
You probably also want:
SOLVNET NO
This turns off the built-in bulk solvent calculation.  If you don't do this, 
refmac will scale its own default bulk solvent mask alongside your 
manually-provided mask.  Then again, you might want to do that just to see what 
happens.  In fact, you can provide refmac with several "partial structure" 
masks.  All of them will get scaled if you put their number on the SCPART line.

If you want the map that refmac uses by default, use the MSKOUT feature to have 
refmac write it out for you.  You can then scale this map with mapmask to make 
sure it has a range from 0 to 1.  You might also want to use other map masking 
tools on it.  Once you have a solvent map you want to use, you can convert this 
map into structure factors using refmac's "mode sfcalc" feature.  Also 
documented on the link above.  Then you can provide them as FPART1 PHIP1 as 
above.

-James Holton
MAD Scientist

On 2/4/2019 4:13 PM, Dorothee Liebschner wrote:
Hi,

Please note that for polder maps, the bulk solvent is reset locally. Turning 
bulk solvent off entirely most likely deteriorates maps.

Best wishes,

Dorothee

On Mon, Feb 4, 2019 at 3:50 AM Samuel Davis (PG Research) 
mailto:s.w.da...@dundee.ac.uk>> wrote:
Hi,

I'm wondering if anyone knows if it is possible to turn off the bulk solvent 
modelling in Refmac5, for the purpose of generating Polder maps? I know that an 
option for Polder maps is directly implemented in Phenix, but we ideally want 
to use Refmac5, as we have used it for the rest of our refinement and want to 
keep it consistent if possible.

Thanks,

Samuel.

Samuel Davis
MRC 3.5 Year Programme PhD Student
Life Sciences, Biological Chemistry and Drug Discovery, University of Dundee
+44 (0)1382 388325 | swda...@dundee.ac.uk

Scottish University of the Year
The Times / Sunday Times Good University Guide 2016 and 2017

Follow my blog: https://musingsofanearlycareerscientist.wordpress.com

The University of Dundee is a registered Scottish Charity, No: SC015096



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1


--
Project Scientist, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
1 Cyclotron Road, M/S 33R0345
Berkeley, CA 94720
Tel: (510) 486-5709
Fax: (510) 486-5909
Web: https://phenix-online.org



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] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Eleanor Dodson]

The keywords are:

SCALE TYPE SIMPLE
SOLVENT NO

I always use for a completed sctructure
SCALE TYPE SIMPLE
SOLVENT YES
and then a set of 4 numbers for scale and Bfactors are output .

There is an option to give these as a FIXED SCALE.

Forgotten the exact keywords but they will be in the manual..
Probably that is the best procedure for each POLDER map?

Eleanor

Maybe t

On Tue, 5 Feb 2019 at 00:13, Dorothee Liebschner 
wrote:

> Hi,
>
> Please note that for polder maps, the bulk solvent is reset locally.
> Turning bulk solvent off entirely most likely deteriorates maps.
>
> Best wishes,
>
> Dorothee
>
> On Mon, Feb 4, 2019 at 3:50 AM Samuel Davis (PG Research) <
> s.w.da...@dundee.ac.uk> wrote:
>
>> Hi,
>>
>> I'm wondering if anyone knows if it is possible to turn off the bulk
>> solvent modelling in Refmac5, for the purpose of generating Polder maps? I
>> know that an option for Polder maps is directly implemented in Phenix, but
>> we ideally want to use Refmac5, as we have used it for the rest of our
>> refinement and want to keep it consistent if possible.
>>
>> Thanks,
>>
>> Samuel.
>>
>> Samuel Davis
>> MRC 3.5 Year Programme PhD Student
>> Life Sciences, Biological Chemistry and Drug Discovery, University of
>> Dundee
>> +44 (0)1382 388325 | swda...@dundee.ac.uk
>>
>> Scottish University of the Year
>> The Times / Sunday Times Good University Guide 2016 and 2017
>>
>> Follow my blog: https://musingsofanearlycareerscientist.wordpress.com
>>
>> The University of Dundee is a registered Scottish Charity, No: SC015096
>>
>> --
>>
>> To unsubscribe from the CCP4BB list, click the following link:
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>>
>
>
> --
> Project Scientist, Molecular Biophysics and Integrated Bioimaging
> Lawrence Berkeley National Laboratory
> 1 Cyclotron Road, M/S 33R0345
> Berkeley, CA 94720
> Tel: (510) 486-5709
> Fax: (510) 486-5909
> Web: https://phenix-online.org
>
> --
>
> 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] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Dorothee Liebschner]

Hi,

Please note that for polder maps, the bulk solvent is reset locally.
Turning bulk solvent off entirely most likely deteriorates maps.

Best wishes,

Dorothee

On Mon, Feb 4, 2019 at 3:50 AM Samuel Davis (PG Research) <
s.w.da...@dundee.ac.uk> wrote:

> Hi,
>
> I'm wondering if anyone knows if it is possible to turn off the bulk
> solvent modelling in Refmac5, for the purpose of generating Polder maps? I
> know that an option for Polder maps is directly implemented in Phenix, but
> we ideally want to use Refmac5, as we have used it for the rest of our
> refinement and want to keep it consistent if possible.
>
> Thanks,
>
> Samuel.
>
> Samuel Davis
> MRC 3.5 Year Programme PhD Student
> Life Sciences, Biological Chemistry and Drug Discovery, University of
> Dundee
> +44 (0)1382 388325 | swda...@dundee.ac.uk
>
> Scottish University of the Year
> The Times / Sunday Times Good University Guide 2016 and 2017
>
> Follow my blog: https://musingsofanearlycareerscientist.wordpress.com
>
> The University of Dundee is a registered Scottish Charity, No: SC015096
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1
>


-- 
Project Scientist, Molecular Biophysics and Integrated Bioimaging
Lawrence Berkeley National Laboratory
1 Cyclotron Road, M/S 33R0345
Berkeley, CA 94720
Tel: (510) 486-5709
Fax: (510) 486-5909
Web: https://phenix-online.org



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1

Re: [ccp4bb] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Edwin Pozharski]

It should be according to the manual

http://www.ccp4.ac.uk/html/refmac5/keywords/xray-principal.html#solv

If you using CCP4i, I believe this is done by unchecking the "Calculate the
contribution from the solvent region" box in Scaling section.

---
I don't know why the sacrifice didn't work. The science seemed so solid.
Julien XIII, Lord of the Lemurs

On Mon, Feb 4, 2019 at 6:50 AM Samuel Davis (PG Research) <
s.w.da...@dundee.ac.uk> wrote:

> Hi,
>
> I'm wondering if anyone knows if it is possible to turn off the bulk
> solvent modelling in Refmac5, for the purpose of generating Polder maps? I
> know that an option for Polder maps is directly implemented in Phenix, but
> we ideally want to use Refmac5, as we have used it for the rest of our
> refinement and want to keep it consistent if possible.
>
> Thanks,
>
> Samuel.
>
> Samuel Davis
> MRC 3.5 Year Programme PhD Student
> Life Sciences, Biological Chemistry and Drug Discovery, University of
> Dundee
> +44 (0)1382 388325 | swda...@dundee.ac.uk
>
> Scottish University of the Year
> The Times / Sunday Times Good University Guide 2016 and 2017
>
> Follow my blog: https://musingsofanearlycareerscientist.wordpress.com
>
> The University of Dundee is a registered Scottish Charity, No: SC015096
>
> --
>
> 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] Turning off the bulk solvent modelling in Refmac5 to generate Polder maps?

[Samuel Davis (PG Research)]

Hi,

I'm wondering if anyone knows if it is possible to turn off the bulk solvent 
modelling in Refmac5, for the purpose of generating Polder maps? I know that an 
option for Polder maps is directly implemented in Phenix, but we ideally want 
to use Refmac5, as we have used it for the rest of our refinement and want to 
keep it consistent if possible.

Thanks,

Samuel.

Samuel Davis
MRC 3.5 Year Programme PhD Student
Life Sciences, Biological Chemistry and Drug Discovery, University of Dundee
+44 (0)1382 388325 | swda...@dundee.ac.uk

Scottish University of the Year
The Times / Sunday Times Good University Guide 2016 and 2017

Follow my blog: https://musingsofanearlycareerscientist.wordpress.com

The University of Dundee is a registered Scottish Charity, No: SC015096



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=CCP4BB=1