Segmented Subtomogram Averaging (SSTA)

Segmented subtomogram averaging (SSTA) was first introduced by Guyomar et al. In the SSTA workflow, subtomogram averaging is performed along the microtubules segment-by-segment so that the resulting averages reflect the local structural heterogeneity.

In cylindra, SSTA can be easily performed owing to the powerful combination of the polars expression system and the built-in methods for cryo-ET.

Note

The EMPIAR deposition by the authors such as EMPIAR-11253 is very useful to test this workflow.

Preparation

After loading the tomogram and the microtubules are properly fitted with splines, molecules can be initialized along each spline at dimeric intervals. The interval can be described by the polars expression col('pitch') * 2, which is evaluated with the "pitch" value of spline global properties.

ui.infer_polarity(splines="all")
ui.global_cft_analysis(splines="all")  # to get lattice pitch
ui.map_along_spline(
    splines="all",
    molecule_interval="col('pitch') * 2",
    prefix="Center",
)

You can further refine the molecules by aligning them to the pre-determined structure of microtubules such as EMD-7974.

Note

The template image must be rotated in the correct direction so that the axis of the template microtubule matches the y-axis.

ui.sta.align_all(
    layers=['Center-0'],
    template_path="path/to/template.mrc",
    mask_params=(0.3, 0.8),
    max_shifts=(1.0, 1.0, 1.0),
    rotations=((0.0, 0.0), (10.0, 2.0), (0.0, 0.0)),
    cutoff=0.2,
)

By the standard subtomogram averaging using average_all, you can check if the alignment was successful.

ui.sta.average_all(layers=['Center-0-ALN1'], size=48.0)

SSTA

Molecules can be grouped by its features. This means that you can obtain averages for each group of molecules. The grouped averaging is a generalization of what SSTA supposed to do. What we need here is to correctly label the molecules for each segment, that is, a column with values like [0, 0, 0, 1, 1, 1, ..., n-1, n-1, n-1] is needed supposing that there are n segments.

Molecules generated by map_along_spline are already labeled with the nth column, which takes the value of [0, 1, ..., N-1]. Expression col('nth') // Ns will group the molecules into segments of Ns molecules. Since we initialized molecules at dimeric intervals, the intervals are usually ~8.2 nm. For 100-nm SSTA, the Ns value should be \(100 / 8.2 \approx 12\).

ui.sta.average_groups(
    layers=['Center-0-ALN1'],
    size=48.0,
    by="col('nth') // 12",
)