@Bassem people often ask questions like this - “do you think X might work given these caveats?” My answer is almost always that I don’t know, but that trying will only cost you a few hours of GPU time. Just go for it!
Of course, I have lots of opinions about what parameters to use, like in the post above.
To add: in relion 4 under optimization you can select to run the initial model in C1 and then ask relion to align to C2 so that the upcoming step is more straight forward in symmetry relaxation. Earlier version of relion you will need to use command line relion commands to set the structure to C2 symmetry.
I gave the masking a go on cryosparc but it did not help. Here is the plan, I will mask both mickey mouse ears of my structure and try that combined with the symmetry relaxation in relion.
Cool, let us know how it goes! The structures need to be aligned with the (pseudo) C2 axis as the Z axis of the box - like the output of a C2 imposed refinement.
Dear All,
I have the same problem.
We have a dimeric enzyme (C2) in complex with substrates and I would like to highlight differences in the active sites of each monomer. How do I do that in practice ?
For the moment, I managed to extract the asymmetric unit, I did relion_particle_symmetry_expansion and I am trying some 3D classification. Can you give me some advice ?
Thanks a lot,
GIA
Hi Gia,
We have a dimeric enzyme (C2) in complex with substrates and I would like to highlight differences in the active sites of each monomer. How do I do that in practice ?
The easiest way would be to symmetry expand, and then perform 3D classification with a focus mask covering one protomer (or perhaps a sub-region of the protomer if you have some prior knowledge about where conformational changes/heterogeneity might be expected).
In practice, how successful this is will depend on the scale of the heterogeneity. If the changes are just binding/unbinding of a ligand, or movement of a small loop, this may be challenging to resolve, whereas if they are more substantial or less localized, they should be easier to detect.
Hi !
Thanks a lot for you advice.
I am new in SPA and for the moment I only know how to do things in Relion (opss…), but I am going to try Cryosparc too !
So If I want to try what you are suggesting, I will:
Thanks a lot !
GIA
Hi Gia,
Same in relion as in cryosparc really:
If the interface between the protomers is somewhat flexible and the monomers are sufficiently large (~>50kDa), it may be worth performing a local refinement with a soft mask over one protomer immediately after symmetry expansion (prior to further classification).
The classification I am referring to is all without alignment - I think that is your best bet with a small mask after symmetry expansion, although you could try with local searches in relion also.
Hi again,
sorry I have trouble understanding the correct way to do this.
When I do 3Dclassification of the symmetry expanded particles, should I use a monomer for the reference AND for the mask ? what will be the output of such calculation ? differences at the level of the monomer, regardless wheter is subunit A or B ?
Also what you mean for “reconstruct” ?
Sorry for the naive questions and thank you,
GIA
When I do 3Dclassification of the symmetry expanded particles, should I use a monomer for the reference AND for the mask ?
No - dimer reference, monomer mask.
what will be the output of such calculation ? differences at the level of the monomer, regardless wheter is subunit A or B ?
Yes.
Also what you mean for “reconstruct” ?
Reconstruct - backproject to generate a volume, which at least in cryosparc is worth doing after classification as the classification is calculated (depending on the target resolution) in a much smaller box than the original box size of the particles.