[buster-discuss] Anisotropic ADP only for selected atoms

Mon Oct 30 10:37:53 CET 2023

Dear ClAuS & Gerard,

Thanks for your answers.

I do have anom signal in my data that also helps me to locate the Cadmium. This dataset was collected at Soleil synchrotron at a wavelength of 1.0 Angstrom. I have others datasets that I collected at 1.5 Angstrom to have stronger anom signal. My resolution is also decent, around 2 Angstrom. The main problem is that for some metal ions I failed to define a first coordination sphere that has chemical sense (I attached a few screenshots). Most of the time is an educated guess by placing waters around.

I tried to run with " refine -M ADP -Gelly ADP-CD.gelly -TLS -autoncs  -nbig 10 InitialiseBiso=50  AutomaticFormfactorCorrection=yes ... '' as you suggested and it worked. Unfortunately there is no improvement in the problematic regions and additionally R free increased significantly from 0.248 to 0.265.

I am afraid I have to live with that. In any case thanks a lot for your advices, it's good to know we have this possibility. At this point I am not trying with SHARP, this would have been an elegant way as well.

Best,
Giacomo

On 27/10/2023 13:12, "ClAuS Flensburg" <claus at globalphasing.com <mailto:claus at globalphasing.com>> wrote:

Dear Giacomo,

There is a mechanism to select specific atoms for anisotropic ADPs.
The default set is "nonHydrogen" - see the list of automatically
determined sets in $BDG_home/tnt/data/sets.dat and how BUSTER is
interpreting them for your specific model in LIST.html (Search for
"set definitions"). If you want to make all CD atoms have ADP's try

cat > ADP-CD.gelly <<EOF
NOTE BUSTER_ANISO_SET { ATOMTYPE CD }
EOF

See also [1] about sets.

Then run BUSTER with

refine -M ADP -Gelly ADP-CD.gelly ...

Does this work for you?

If your data contains a good anomalous signal and the input file
contains them BUSTER will write F_ano/PHI_ano columns that are
automatically consumed by COOT to display anomalous maps.

Hope this helps.

With best wishes,

ClAuS & Gerard (for BUSTER Developers)
[1] https://urldefense.com/v3/__https://www.globalphasing.com/buster/manual/gelly/manual/gelly2.html__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4QzwBZDE$ <https://urldefense.com/v3/__https://www.globalphasing.com/buster/manual/gelly/manual/gelly2.html__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4QzwBZDE$> 

On Thu, Oct 26, 2023 at 09:51:31PM +0100, Gerard Bricogne wrote:
> Dear Giacomo,
> 
> Thank you for your enquiry. 
> 
> At the moment we do not have such a selection mechanism, but let me
> suggest another approach. If your data were collected with a typical
> wavelength of about 1.0 Angstroem, Cadmium has an f" slightly above 2
> electrons, so you should be able to see them in an anomlous difference
> Fourier map. You could then use SHARP, with phases from your model given as
> external phase information, to refine ADPs for these "heavy atoms. Finally,
> you could then add these ADPS to the same Cd atoms in the PDB file of your
> model.
> 
> Just a thought, with possibly a number of devils hiding in the details
> but with perhaps a chance of success.
> 
> What is the resolution of your data? Where do you do your data
> collection?
> 
> 
> With best wishes,
> 
> Gerard.
> 
> --
> On Thu, Oct 26, 2023 at 05:28:39PM +0000, Giacomo CARLONI wrote:
> > Dear Buster users,
> > 
> > I was wondering if there is a way to specify a subset of atoms to be refined with anisotropic ADP, while keeping the rest isotropic (1 TLS per chain).
> > 
> > I have a dataset at 2.0 A resolution with more than 30 Cadmium ions coming from the crystallization condition. The majority of them is not at full occupancy and often the best I can do is placing random waters nearby. I would like to test if I can improve the situation by treating the metal ions anisotropically.
> > 
> > Thanks for your help,
> > 
> > Giacomo
> 
> > _______________________________________________
> > buster-discuss mailing list
> > buster-discuss at globalphasing.com <mailto:buster-discuss at globalphasing.com>
> > https://urldefense.com/v3/__https://www.globalphasing.com/mailman/listinfo/buster-discuss__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4Md-LzEk$ <https://urldefense.com/v3/__https://www.globalphasing.com/mailman/listinfo/buster-discuss__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4Md-LzEk$> 
> _______________________________________________
> buster-discuss mailing list
> buster-discuss at globalphasing.com <mailto:buster-discuss at globalphasing.com>
> https://urldefense.com/v3/__https://www.globalphasing.com/mailman/listinfo/buster-discuss__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4Md-LzEk$ <https://urldefense.com/v3/__https://www.globalphasing.com/mailman/listinfo/buster-discuss__;!!JFdNOqOXpB6UZW0!sshJ9RsAksAntdoPUIdy1_BjdkXGE5pFS2WdX9KQUz0rc-87K_VuK26IyVKT14HGA5YxiIg8KRixbAo7YPm4Md-LzEk$> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2023-10-27 at 14.53.40.png
Type: image/png
Size: 446854 bytes
Desc: Screenshot 2023-10-27 at 14.53.40.png
URL: <http://www.globalphasing.com/pipermail/buster-discuss/attachments/20231030/36d04706/attachment-0003.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2023-10-27 at 14.48.28.png
Type: image/png
Size: 643299 bytes
Desc: Screenshot 2023-10-27 at 14.48.28.png
URL: <http://www.globalphasing.com/pipermail/buster-discuss/attachments/20231030/36d04706/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Screenshot 2023-10-27 at 14.43.43.png
Type: image/png
Size: 326245 bytes
Desc: Screenshot 2023-10-27 at 14.43.43.png
URL: <http://www.globalphasing.com/pipermail/buster-discuss/attachments/20231030/36d04706/attachment-0005.png>