Multi-particle refinement

See also

We highly recommend the M paper to better understand what this software does and is capable of.

Now that our particle poses are mostly correct and we have a good estimate of the reconstructed density, there are several parameters that can be tweaked a posteriori to further refine our result. We will perform these optimisations in M.

Create Population and Species

A project in M is referred to as a population. To create a new one press the big + button in the center of the window, and provide a directory. Here, we put it in root/M/1.6Apx/.

We then need to add a data source. Click on Manage data sources to add a local source, and select the *.settings file inside the frames directory. Make sure to Include items outside of filter ranges (since we did not use the filters for this project), give it a name such as HIV and press CREATE. We also only want to Use only first 15 frames/tilts in order to refine only based on the high resolution information.

We can then create a Species by pressing the next big ‘+’, selecting from scratch and filling in the parameters as follows:

Afterwards, select the two last halfmaps generated by the RELION refinements at 1.6Å/px. For the mask, we will generate a binary mask around the central hexamer (160 Å diameter).

Note

Note that this time we are not adding a soft edge: M wants a binary mask!

Mask one of the halfmaps to include only the central hexamer in Dynamo then use your favourite program and determine an appropriate threshold for binarisation. Use this as the --ini_threshold parameter in the following relion_mask_create command, replacing <latest_halfmap> as appropriate:

relion_mask_create --i <latest_halfmap>.mrc --angpix 1.6 --extend_inimask 5 --o mask_1.6Apx.mrc --ini_threshold 0.05

If using Relion3.1+, we need to reformat the final particle positions .star file using relion_star_downgrade to the Relion3.0 specification (Warp does not support Relion 3.1+):

relion_star_downgrade -s run_it019_data.star

M will only look for file names with the pattern *_data.star, so we need to rename the output from relion_star_downgrade (e.g., from run_it019_data_rln3.0.star to rln3.0_run_it019_data.star). Now, select this reformatted .star file containing the final particle positions from Relion as the input for particle poses in M. Click FINISH to generate the new species.

Refinement Parameters

To start the refinement, click on the REFINE button and fill in the parameters as follows.

Note

We will do a few iterations of refinements: first some with only spatial parameter optimizations, and then with most parameters. We chose to first optimise a smaller number of parameters to avoid falling into a local minimum, there may be a better way to do this!

First, run four iterations with these parameters:

Afterwards, three iterations as follows:

End Result!

If everything went well, after the second round of refinements you should be able to get to something like this:

Here’s a closer look to the map compared to the available structures:

Congratulations with your reconstruction!