Hi,

yes, as long as the map shows the glycans clearly enough so that you can model them, then there is no problem refining your model using phenix.real_space_refine. Normally, there is no need to add anything special.
Now I'm puzzled.. do you have a reason to believe that something isn't working correctly?

By the way, we will be at Tsinghua University first week of September running Phenix workshop. We can discuss things while there!

All the best,
Pavel

On 8/7/18 08:16, Miao Gui wrote:
Dear phenix developers,

I'm a PHENIX user. PHENIX helped me a lot when refining structural models against cryo-EM maps.

Recently, I built a structural model of a glycoprotein and added the glycans using COOT against a 3.4 A cryo-EM map. Then I want to refine the model using PHENIX.real_space_refine. I always add the second structure restrains when refining the models against cryo-EM maps. Do I need to add other restrains for the glycans? Is it appropriate to refine the glycoprotein using the real_space_refine?
Thank you.

Best wishes,

Dr. Miao Gui
Tsinghua University, Beijing