Skip to content

Relion post processing

Jump to bottom
Michael A. Cianfrocco edited this page Jun 7, 2018 · 1 revision

Home > Using Relion for data processing > Relion post processing

##Relion post processing

During 3D auto refine in Relion, Relion calculates the FSC between two independent half maps ('gold standard FSC') to estimate resolution. But, since it is doing this resolution calculation using a circular mask based upon the diameter input by the user, this resolution estimate is an underestimate.

Therefore, Relion has a post processing step, which will perform the following operations:

  • Combine independent half datasets into a single map
  • Mask your structure using a tight mask
  • Reweight the 3D information by applying a negative B-factor
  1. Inputs required for post processing
  2. Launching and using Relion's GUI
  3. Post processed map outputs

###Inputs required for post processing

The following inputs are necessary for Relion post processing:

  • TWO final half maps from 3D auto refine
    • These half maps are produced during the final step of Relion 3D auto refine
  • MTF of detector (K2, Falcon, etc.) in .star format
  • Estimate of threshold for creating a map around 3D volume
    • Open one of your half maps and display the volume using Chimera, adjusting the threshold to a point where all of the density is shown (no floating satellite densities). Use this number for post processing

###Launching and using Relion's GUI

$ relion &

Which will open this window:

  • Input pixel size for your data. This will need to be adjusted to a new pixel size if you binned your data during extraction
  • The diameter selected must extend beyond the protein density that you expect, but if it is too wide, you may include too much information and lead to artifacts during analysis.

Now, click on the *Post-processing tab on the left hand list. You will need to enter the following inputs:

  • One of your half maps from the final step of 3D auto refine
  • Rootname for output - for clarity, it helps to name this output volume the same basename as the half map from 3D auto refine

Clicking onto the next tab, Mask, select auto masking and provide the following:

  • The threshold at which all of your density is contained within your map (using Chimera)
  • Extend binary range and add soft edge - this will make a mask that is slightly bigger than the thresholded mask using the value provided above. Typical values are 3 - 5 pixels to increase the mask, with a 2 - 3 pixel soft edge

Clicking on to the next tab, Sharpen, you will provide MTF information for your detector:

  • MTF of detector - This is a .star file for the MTF of your detector. Links for MTFs: Falcon-II at 300 kV and K2 Summit at 300 kV
  • Estimate B-factor automatically - If your structure is higher resolution than 15 Angstroms, Relion can estimate the B-factor of your data. Otherwise, you will need to provide an ad-hoc B-factor for your data. Typical values could be 200 - 600.

Clicking on to the next tab, Filter:

  • Skip FSC-weighting - You should only skip this input option IF your map has local resolution regions that are higher resolution than the average FSC of the whole map. By providing a higher resolution filter, you can visualize higher resolution features of your structure (as long as they exist)

And this point, you can print the command and it will look something like this:

*** The command is:
`which relion_postprocess` --i Refine3D/run1_iter025 --o Refine3D/run1_iter025_half1_class001_postprocess  --angpix 1 --auto_mask  --inimask_threshold 0.02 --extend_inimask 3 --width_mask_edge 3 --mtf k2_summit.star --auto_bfac  --autob_lowres 10

###Post processed map outputs

You can run this command directly on your workstation/log-in node because it will take only a few seconds/minutes.

Make sure to take note of the final information that Relion tells you regarding the final resolution of your 3D map.

The outputs will be the following:

  • Refine3D/run1_iter025_half1_class001_postprocess_automask.mrc - Binary mask generated by Relion to mask your half volumes
  • Refine3D/run1_iter025_half1_class001_postprocess_fsc.xml - FSC curve in XML format, ready for upload to EMDB
  • Refine3D/run1_iter025_half1_class001_postprocess_masked.mrc - Masked input map
  • Refine3D/run1_iter025_half1_class001_postprocess.mrc - Final, reweighted, filtered map for structure interpretation
  • Refine3D/run1_iter025_half1_class001_postprocess.star - .star file containing information about your 3D map (FSC, etc.)

You can plot the FSC curve of your corrected FSC map by plotting in gnuplot:

$ relion_star_plottable {postprocess star file}.star data_fsc _rlnFourierShellCorrelationCorrected _rlnResolution

This will open a gnuplot window showing your FSC curve, plotted with corrected FSC on Y-axis and spatial frequency on the X-axis:

Home

Clone this wiki locally