On 02/07/2024 13:17, Noha Elhosseiny wrote:
--
Thank you very much.
I am trying to build a whole C15 model because I usually undergo
refinement in Phenix after a round of fitting in Coot. What I have
done so far was do one chain, copy and superpose, then use this model
and my map for real space refinement. I guess the problem is that the
map coordinates are not exactly correct.
Noha
On Mon, Jul 1, 2024 at 7:20 PM Paul Emsley <pems...@mrc-lmb.cam.ac.uk>
wrote:
On 7/1/24 12:24, Noha Elhosseiny wrote:
Thanks Paul,
I replied below in a different color
On Mon, Jul 1, 2024 at 2:17 PM Paul Emsley
<pems...@mrc-lmb.cam.ac.uk> wrote:
On 01/07/2024 11:59, Noha Elhosseiny wrote:
>
>
> --
>
>
>
> I am new to coot and structural analysis in general. I am
trying to do
> model building for a symmetric protein consisting of 15
identical
> chains, using the density map from cryoEM analysis. I made
an initial
> model by first fitting one chain (chain A- roughly) in
Chimera, copied
> this chain to make the 15-mer, and then moved to coot for
refinement
> and improving the fitting. I noticed that there is a slight
shift from
> the best fit of the chain I fitted and then copied (chain
A) in the
> other chains in the density map. To cut it short chain A is
perfect,
> but chains B-O gradually shift very slightly from the ideal
fit,
> probably as a result of the copying process. Now I am not
sure what to
> do. I can try to work on each chain individually to fall
into the
> density like I did with chain A, but it seems
counterintuitive to do
> this 14 times when I already have a perfect chain.... I
considered
> doing LSQ superpose of chain A on each of the other chains,
but given
> they are already shifted...I think this will not solve the
problem. I
> also think I can try to do "refine chain" or "jiggle fit",
but I think
> this will make it worse not better as the chains are very
slightly
> shifted.....
>
> Any suggestions on how to proceed?
>
Presumably you have the transposition matrices from your cryo-EM
processing. I imagine that it might even be proper symmetry.
You can use
that. What molecular symmetry do you have?
I suppose I have...but I do not know how to get these matrices or
where to look for them....I am using CryoSPARC. I can look into
where to get them. I have a C15 symmetry and I am 90% sure it is
correct as the protein is not new but the species is and there
are sequence differences.....
One generally modifies only one copy. When refining, one might
make a molecule of 5 chains (to model the interaction between
chains) and then modify only the middle (C) chain.
You only need do the full symmetry expansion for deposition and
making figures.
Attached is a script to do what you want (as far as I can tell) -
you will need to change the imol file and the centre to be
appropriate for your molecule.
Paul.
> I am trying to build a whole C15 model because I usually undergo
refinement in Phenix after a round of fitting in Coot
So, do you think you need a whole C15 model to refine in Phenix?
I doubt that that is the case (I mean, I would be very surprised if you
needed to do so). I imagine Phenix is able to refine a handful of
monomers just like Refmac or Coot can.
> What I have done so far was do one chain, copy and superpose, then
use this model and my map for real space refinement.
Sounds like quite some palaver.
> I guess the problem is that the map coordinates are not exactly correct
If you are using Coot, that is not a good place to start.
Were you able to find the transformation matrix from the output of
CryoSPARC?
Were you able to run the script to correctly produce your whole C15
molecule?
Paul.
########################################################################
To unsubscribe from the COOT list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=COOT&A=1
This message was issued to members of www.jiscmail.ac.uk/COOT, a mailing list
hosted by www.jiscmail.ac.uk, terms & conditions are available at
https://www.jiscmail.ac.uk/policyandsecurity/
