Hi,
I am sure I did something wrong here
Any Idea how that have happened ?
Hi @sgajew! Welcome to the forum!
This is certainly a strange Orientation Diagnostics result. If you look at the rings in the 3D plot of the Relative Signal (top image, right), you can see the high relative signal is essentially all in rings around the X, Y, and Z axes. In addition, your volume does not seem to have very many features – certainly not what we’d expect to see in a protein around 5 Å.
About how big (in kDa) is your target? How many particles do you have in the final refinement? Can you share some of your 2D classes of these particles?
1 Like
Welcome to the forum. I would second @rwaldo’s request for some 2D classes, because honestly that 3D internal of the spherical assignment looks like nothing - certainly nothing that is 4 Angstrom.
My guess is it’s aligned noise across X/Y/Z, but further information will help diagnose better. 
This is supposed to be a 135kDa complex with a 30kDa component known to be “low density”, somewhat flexible, or occasionally entirely AWOL. We suspect some heterogeneity due to multiple conformations could be present and we tried stacks between 130,000 and 960,000 particles in refinement. This one has 425,000. The data is from a 200 kV accelerating voltage microscope.
EDIT: Maybe I should add that the components of the complex can actually be fitted quite nicely into the, admittedly featureless, blobb.
There is at least something there, but nothing would suggest that it would achieve 4 Å…
I’d suggest heterogeneous reconstruction (multiple rounds) to try cleaning further.
Thanks for sharing those 2D Classes! As @rbs_sci said, you definitely have something there, but the spikey features in your class averages and featureless 3D map make me think it’s not being aligned very well at the moment. I also agree with the recommendation of multiple rounds of heterogeneous refinement – if you’re not familiar with this procedure, we have a case study which covers particle curation in some detail.
Good luck, and please always feel free to come back with more questions!
1 Like