Classify 3D structural heterogeneity

Filling in the GUI

For 3D refinements, select the run-type of 3D reconstruction from the drop-down menu at the top of the GUI.

I/O tab

• See the Prepare input files page on how to prepare your input.

• The pixel size (in Angstrom) should be the same as the one used to estimate the CTF parameters.

CTF tab

• If no CTF correction is to be performed, make sure you phase-flipped your data during preprocessing. See the Prepare input files page.

• Normalisation correction is robust and therefore recommended in the general case.

• Intensity correction corrects for distinct grey-scale intensities among the signal in the data, e.g. because due to distinct SNRs among the micrographs. This option is only effective if the data is provided in a STAR file that contains multiple unique strings for the rlnMicrographName label (see the Prepare input files page.

Optimisation tab

• Often 25-50 iterations are necessary before the refinement converges to a stable solution. Note there is currently no convergence criterion implemented, so the user is responsible for monitoring the convergence.

• The number of classes is the most important parameter. Often one performs multiple calculations with different values.

• The regularisation parameter determines the relative weight between the experimental data and the prior. Bayes' law dictates it should be 1, but better results are often obtained using slightly higher values (e.g. 2-4), especially when dealing with cryo-data.

• The particle diameter (in Angstroms) serves to define a soft spherical mask that will be applied to the experimental images to reduce their background noise. If solvent flattening is set to Yes, then also the references will be masked using the same spherical mask (or using a user-provided one under the solvent mask option).

Sampling tab

• In-plane angular sampling rates of 5 degrees are enough for most applications.

• Translational search ranges may depend on how well-centered the particles were picked, but often 10 pixels will do the job (translational searches in subsequent iterations are centered at the optimal translation in the previous one, so that particles may "move" much more than the original search range during the course of an entire refinement. Note that pre-centering prior to RELION refinement is not necessary, and also not recommended (it often messes up the Gaussian distribution of origin offsets).

Running tab

• It is unlikely one needs threads for 2D class averaging, as this typically takes only modest amounts of memory. Therefore, in case multiple CPUs are available for this task, use the more efficient MPI parallelisation.

Continuing an old run

If after completion of a run one decides the refinement had not yet converged, or if one wants to perform additional iterations with for example a finer angular sampling, one can continue an old run by selecting the corresponding option from the drop-down menu at the top of the GUI. The number of iterations of the new run should be higher than the iteration from which to continue, as the iteration numbering will continue from the latter. Note that some options cannot be changed upon continuing an old run, and these are deactivated in the GUI.