Hi everyone,
I would like to use PHENIX for refinement of a cryoEM map of an icosahedral virus particle that I generated using EMAN2.
For this, I converted the .mrc file of my EM map to an .mtz file – this served as the input map file. A .vdb file of the wild type virus (I am solving structure of the mutant) is available on Viperdb, and I used this as the model file.
While setting a real space refinement, I'm getting an error of duplicate
atom labels:
“Sorry: number of groups of duplicate atom labels: 7671
total number of affected atoms: 460260”
How can I circumvent this problem?
Also, any suggestions on how to perform refinement of cryoEM maps of virus
particles (i.e. repeating icosahedral asymmetric units) using Phenix would be very helpful.
Dear Saumya,
I'm not entirely sure whether I understand your Question correct.
You aim to do a real space refinment in PHENIX?
If so, why do you convert your Map into reciprocal space?
You can use Phenix.real_space_refine on your cryo-EM map directly.
The Error you report might be caused by a duplication of chain IDs (two times chain A for example).
I hope this helps you.
Good Luck
Thiemo
Thanks for you input, Thiemo!
I have an mrc file of my map - and I'd read that the input files need to be in either ccp4 or mtz format, so I had converted it to mtz.
Since I have a map of a virus capsid, it is made up of multiple copies of the same protein, hence the duplication of chain IDs. What do I do in such a scenario?
Ok, but if you like to try real space refinement on your experimental map
PHENIX can work on mrc files, I always link a .mrc file to a .map file and use this for PHENIX.
Regarding the Chain ID's you have to give every chain a unique ID. You can use a-z, A-Z and 0-9. You can change the IDs either in COOT or in a text editor manually or try to use Chimera, copy combine, rename them uniquely.
Maybe their is a better way for symmetrical Models but I never work on a symmetric model so far.
Good Luck!
Hi Saumya Bajaj , mtz file generally denotes fourier coefficients for the map (cryo-EM) while it represents the diffraction data for X-ray crystallographic maps.
Both of them may be used for fourier/reciprocal space refinement. For cryo-EM, we generally prefer real space refinement as discussed above.
Hi Saumya,
I read your question and facing same issue.
How did you overcome duplicate atom error and merged your symmetric PDBs?