Zoom Transcript

---

Would we see the same alternate conformations for the same protein structure solved by X-ray diffraction and by Cryo-EM at the same resolution?

---

Which restraints help to keep the geometry of secondary structures intact in phenix rspace refinement at 3.5 to 4.0 Ang resolution range?

---

Secondary structure restraints are designed to keep the geometry of SS good at low resolutions. They are turned on by default at 3.5-4A resolutions in real-space refine.

---

Are there limits the PDB sets for how “bad” validation statistics can be? For example... clash score 70?? favored rama less than 70%?? It seems to me statistics that poor should be unacceptable, nature and otherwise.

---

@Jonathan, it works another way around. If a journal is fine with publishing the paper, PDB is accepting the structure. PDB does not choose to accept/reject a submission.

---

I see... thank you, Oleg

---

i have a cryo-EM density of a dimer at 2.8A, however to improve density of one of the domain i have to partially subtract signal creating a monomer and split it into 2-3 rigid bodies. Now i can build all domains. I can now create a composite map and do refinement of the whole structure. is that a good strategy? or should i do it in a certain order: starting from individual domain, then monomer and then a dimer?

---

Is there a way to check

---

-if scale of a map (originating from say error in pixel value), is accurate, and if so, how to fix. Can a minor error create major complications for refinements of map in the resolution ranges of 2.5-4 amgstrom

---

i’d also like to profit from density modification in phenix

---

Thank you - i will have to explore the best strategy.

---

Thank you so much Pavel

---

@Pavel, For less experienced users (like myself), when do you stop fiddling with parameters and just accept the refinement results? E.g. my first run (during this session) resulted in RamaZ of 1.47 (Ramachandran allowed/favored - 2.9 and 97.1), second run with all run options enabled gave RamaZ of 0.2 and (3.4 and 96.56 for allowed and favored). Which one should I chose? Strangely for the last run, when i do validate Ramachandran plot in COOT, it tells me there are 2 outliers..

---

What’s the best way to make restraints for ligands like ADP:Mg:H2O’s?

---

thanks.

---

thanks!

---

Thank you very much for a great talk Jane. I wonder what could be the physical “basis” behind the few outlier clashes typically found in any accurate structure? (given that two non-bonded atoms cannot overlap…?)

---

Some time the ligand geometry is out and is very difficult to fix. Is there a way for a ligand to fix it easily and with few refinement run. Thanks

---

Thank you!

---

@Pavel or Nigel; is there any functionality in Phenix equivalent to prosmart restraints in CCP4?

---

thanks

---

Thank you appreciated!

---

What format is the best for input files for creating a cif file for a small molecule?

---

SMILES, PDB, MOL2?

---

In structure-based drug discovery, we often create new compounds to fit in electron density

---

Thanks

---

!

---

Thank you!

---

Thank you very much for the workshop :)

---

Thanks to all the speakers and organizers!

---

Thank you for organizing this!

---

Thank you very much!

---

Thanks!