I’m working on a protein complex composed of three proteins.After the ab-initio reconstruction I used particles_all_classes and volume_class_0 to run the homogenous refinement. I put the symmetry as default(i.e C1) and with this, I have got 6.5A resolution cryo-EM map. I could see two protein molecules but I am not sure how to assess the quality of the map and which step would be best to proceed next. Also, can model building be done with such low-resolution map?
Any advice on this would be helpful.
Thanks a lot.
the universal problem in cryoEM is that we are trying to study proteins/complexes that are heterogeneous. they are moving, sometimes the complex is not formed or a binding partner is missing, some particles may be damaged while others are not, etc. The goal is to remove all particles that are bad, and then classify all good particles into conformationally and compositionally homogeneous groups. So all are in the same position and contain the same protein(s). For this, I suggest heterogeneous refinement as well as 3D variability analysis. 3D classification may be used later, but requires some experience. Absolutely for you I would start with tutorials available online, you can download datasets from EMPIAR and follow instructions to learn how and when/why to classify particles in 2D and 3D.
It is possible to learn some things from your low res map, but you have to consider it is actually the combination of several high-res structures which are incorrectly overlayed on each other. so best to separate the things that are alike and get to higher resolutions before modeling.
Fast track: take the output of refinement - the particles and 4 copies of the 6.5Å map - into heterogeneous refinement and let it sort them out a bit more. then do homogeneous refinement and/or Non-uniform refinement on each of the classes - So should be only 1/4 the amount of particles or so…
Thank you for your answer.
yes, it’s a bit difficult to understand the sample if it is heterogenous and if the complex is unstable/damaged.
I’d look into more tutorials to get more ideas about classifying particles in 2D and 3D. Also, try to follow the fast track option.