Dear Sjors,
Since you fit the result in Chimera, why don’t you directly simulate the map in chimera?
Any issues doing that, since it is quite straightforward?
Best,
Amedee
> On Jan 26, 2016, at 2:22 PM, Sjors Scheres <[log in to unmask]> wrote:
>
> Dear Pierre-Damien,
>
> We usually use EMAN2 (e2pdb2mrc.py) to make mrc files from PDBs. But as
> far as I understand it (Steve may hopefully point out that I'm wrong!) the
> origin gets displaced with the generated mrc maps. We then usually use a
> quick fit-in-map from UCSF Chimera (and then 'vop resample onGrid) to
> replace the pdb-derived on top of the original reconstruction. Then, you
> should indeed expand this mask and put a soft edge on it (we use
> relion_mask_create for that). Finally, just run the postprocessing step
> from the GUI with the mask to get a resolution-estimate for that
> region-only. You can do all these steps for all different proteins in your
> complex. Actually, it is often very useful to have differently-filtered
> maps when you're building and refining the atomic models anyway.
>
> I think ResMap is a sound alternative to the local-resolution estimation
> problem. Note you can conveniently run it from the GUI using the same
> half-maps you would use for the postprocessing. It usually doesn't take
> very long to run and you don't need a cluster for it. In this case, always
> provide a mask around the entire complex (as for example obtained in the
> postprocessing step).
>
> As a reviewer casted doubts on the actual resolution numbers generated by
> Resmap, we compared both approaches last year for different regions of the
> tri-snRNP spliceosomal complex, see the supplementary material of Nguyen
> et al, Nature 2015. You'll see that the results are fairly consistent, so
> either approach would be OK.
>
> HTH,
> Sjors
>
>
>> Hello everyone,
>> I would like to know what is the procedure to compute the resolution
>> (from the fsc curve) for each factor within a complex.
>>
>> If I understood correctly, I made a plan but I have some unanswered
>> questions :
>> -you generate a mask from your pdb protein model (cp from pdb in spider
>> or fit in map using chimera with the generation of a map filtered to X
>> resolution)
>> -you will have to pad your mask, expand it, threshold it and align it
>> with your initial volume
>> -multiply your mask to your two "half" maps (generated by relion refine
>> command)
>> -compute the fsc curve with the masked "half" maps
>>
>> Is there an easy way to do all this without using several programs ?
>> What is the exact command in relion to compute only the fsc without
>> running a refinement ?
>>
>> Another alternative would be to use resmap. But how do you EASILY
>> generate a map from the pdb model which can have the same dimensions as
>> your EM map ? And it seems to me that resmap is much more time consuming
>> than the first solution...
>>
>> Thank you in advance for your help
>>
>> Pierre-Damien
>>
>
>
> --
> Sjors Scheres
> MRC Laboratory of Molecular Biology
> Francis Crick Avenue, Cambridge Biomedical Campus
> Cambridge CB2 0QH, U.K.
> tel: +44 (0)1223 267061
> http://www2.mrc-lmb.cam.ac.uk/groups/scheres
|