How to combine two local-refined maps of two parts from one protein into one map in Cryosparc?

Local Refinement
1

Hi, all,

I manully seperated my protein into four parts and locally refined each by using the subtracted particles and it indeed give us some better results. However, this will lead to recenter each part at the center of box, since the defined fulcrum setting. Now I want to combine four local-refined maps into one map for modelling. Could you tell me whether it is possible or not? How to handle this in Cryosparc or any others?

thanks!

Niu

1 Like
2

Hi,

In the “new” local refinement in 3.1, each part remains in its original location - not recentered in the middle of the box. This makes map combination and other downstream operations such as particle subtraction more convenient.

For map combination, aligning maps to a global reconstruction and combining by taking the maximum value at each voxel (vop maximum in UCSF Chimera) is the best method I have found to get a seamless combination. I would still encourage deposition of the orginal component maps as well, though.

Cheers
Oli

5 Likes
How do I reconstruct my whole protein after local refinement
Local resolution colored map with composite maps
3

Hi @niu,

Seconded Oli’s suggestions. Using the new version should preserve the original orientations and positions of each body/mask, making it easier to combine maps using the procedure Oli mentioned. I don’t know too much about what the EMDB standards are, but it appears that individual body maps are routinely uploaded (e.g. see entry 21105, from a RELION multibody refinement).

Best,
Michael

4

Thanks for your quick reply. I will try the “new” local refinement later. Also, I agree with you to deposite the orginal maps in the final state. Here actually, the combinated map is just used for modelling the initial structure more easily at the beginning.

5

Thanks a lot. Maybe Relion multibody refinement is another choice and I can have a try later.

6

Hi, Oli,

Let us suppose a situation like this. This dataset has 600K particles, and regular 3D classification did not produce a consensus map in which the four parts, i.e., P1, P2, P3, and P4, are more or less visible at the same time. Therefore, each part‑focused 3D classification and local refinement has been performed, and the four parts can each be seen clearly in their own focused map. However, the maps of the four parts are derived from different particle stacks: P1, 80K; P2, 90K; P3, 110K; P4, 77K. Since restricted masks are used for each part — i.e., with a very small dilation radius and soft padding width being applied — the four local maps hardly overlap. In this situation, can we perform a “vop maximum” (or volume maximum) if no consensus map for the overall structure can be found and local maps are not overlapping? If no, what shall we do?

Look forward to your reply. Thanks.

Lau

7

Hi Lau,

It is a little difficult to visualize - but I would say if they don’t align to the consensus map, I’m not sure that it makes sense to deposit a composite map… perhaps just deposit the local refinements, and show in figures how they relate? Even if you lowpass filter the composite map you cannot align them?

Cheers

Oli