[phenixbb] treating the water molecules in the model

[Pavel Afonine]

Hi Dmitry,

Here are the considerations regarding water picking in cryo-EM maps:

Unlike crystallography, where changing the model changes the maps (e.g., 
adding more structure improves the maps), in cryo-EM, changing the model 
does not change the map. Therefore, tasks such as refining B factors and 
adding water are best done using the most complete and accurate atomic 
model of the macromolecule, preferably towards the end.

Once you have added water, you need to run a round of refinement and 
check all water manually. While automated tools generally add water 
reasonably well, there is no 100% guarantee that all water will be 
correct. Typically, you perform this step once towards the end.


"Flying water" seems to be an issue indeed. While I'm working on the 
solution, a quick and easy workaround for you to use now is to restrain 
water positions to their original coordinates using weak harmonic 
restraints. Keywords you'll need:


reference_coordinate_restraints.enabled=true

reference_coordinate_restraints.selection=water

reference_coordinate_restraints.sigma=0.2


(also can be set up in the GUI).


Let me know if you have any questions or need help!


Good luck!

Pavel


On 1/19/24 03:33, Dmitry Semchonok wrote:
>
> Dear colleagues,
>
> Please share your experience with how you handle the water molecules.
>
> So, we have a refined model and added the water molecules using the 
> Douse protocol.  So, should we refine the model again afterwards? Or 
> is the model considered ready?
>
> We tried to refine our model with water, but then the water molecules 
> flew away from the places.
>
> Please let me know what is the correct way to proceed.
>
> Many thanks for considering my request.
>
> Sincerely,
>
> Dmitry
>
>
> _______________________________________________
> phenixbb mailing list
> phenixbb at phenix-online.org
> http://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe:phenixbb-leave at phenix-online.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://phenix-online.org/pipermail/phenixbb/attachments/20240119/d1cbe21e/attachment.htm>