2D Classification Strategy for Flexible Multi-Domain System

Cryo-EM Data Processing
1

Hi all,

I am relatively new to cryo-EM and would appreciate some guidance on approaching 2D classification for a flexible multi-domain system.

My protein consists of two domains (A and B) connected by a linker. I have structures for domains A and B individually, but the full-length construct (A–linker–B) appears to adopt multiple conformations due to the instrinsic flexibility.

What would be the recommended approach for 2D classification in this scenario? Specifically:

  • Should I focus on one domain initially and mask out the flexible region?

  • Are there particular strategies for handling conformational heterogeneity at the 2D stage?

  • Any recommendations for particle picking given the expected structural variability?

Thanks in advance for any suggestions.

2

2D is used to visualize and characterize if your particle picking has identified appropriate particles, and whether they are conformationally and compositionally homogeneous. I would accept all particles which include either domain A or B and solve the heterogeneity in 3D. What is the purpose of your cryoEM experiment, given you have structures of A and B, and know (or now know) that they are flexibly tethered and you will likely struggle to get an A-B structure? Are you expecting A-B stable interactions?

3

We aim to obtain A-B structure(s). With several low-resolution structures now in hand, the next goal is to identify top semi-stable conformations.