Collapsing

On this page, you'll process the time-series linked_data into collapsed_data which summarizes each microbot into a single row.

Collapse data

The convenient function collapse_data calculates the most common metrics used from the metrics calculated from ImageJ Segmentation. To see the source code, click the blue "source" button.

MicroTracker.collapse_data — Function

collapse_data(linked_data::AbstractDataFrame, translation_dict::Dict)

Collapse each time-series microbot trajectory into a single row of summary data for each microbot.

Uses the translation_dict from Translation Dictionary to include the experimental data.

Output column definitions

V : The mean of the instantaneous speed, dp_um. Always positive, as it is a magnitude. Units of µm/s.
Vx : The mean of the numerical derivative of the x-position. Can be positive or negative. Units of µm/s.
Vy : The mean of the numerical derivative of the y-position. Can be positive or negative. Units of µm/s.
Area_um_mean : The mean of the area of the microbot. Units of µm^2.
Ω_est : The estimated rotation rate extracted from the FFT of the Major_um column in the linked data. Performed using the estimate_omega function. Units of Hz.
R : The bounding-circle radius/radius of gyration. Calculated as the 95th percentile of the major axis Major_um divided by 2. Units of µm.
Circularity : A quantifier based on the aspect ratio of the fit ellipse. Calculated from ImageJ. See their docs. Unitless.
total_displacement_um : The total displacement of the microbot over its entire trajectory. This is already constant in linked_data, so just take one of the values. Units of µm.

Example

julia> collapse_data(linked_data, translation_dict)
37×13 DataFrame
 Row │ particle_unique  filename      V          Vx           Vy           Area_um_mean  Ω_est      R         Circularity  total_displacement_um  B_mT     FPS      f_Hz  
     │ String15         String15      Float64    Float64      Float64      Float64       Float64    Float64   Float64      Float64                Float64  Float64  Int64
─────┼────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────
   1 │ 5_13p5_61p35-0   5_13p5_61p35  88.7683    73.1812      -0.533202       524.844    0.0866525  28.8823      0.466515            425.447         13.5    61.35      5
   2 │ 5_13p5_61p35-1   5_13p5_61p35  84.2168    75.1932      -1.43098        723.553    0.184789   37.7608      0.447464            204.647         13.5    61.35      5
   3 │ 5_13p5_61p35-2   5_13p5_61p35  34.589     25.285        3.95886         66.7345   4.94208    12.1932      0.706898            151.268         13.5    61.35      5
   4 │ 5_13p5_61p35-3   5_13p5_61p35  61.7119    51.8247      -3.01578         80.8338   5.06137    14.7728      0.647094            170.821         13.5    61.35      5
   5 │ 5_13p5_61p35-4   5_13p5_61p35  20.6538     7.68584      0.0244971       23.9995   2.51434     2.91041     0.984024             15.7076        13.5    61.35      5
   6 │ 5_13p5_61p35-5   5_13p5_61p35  21.9972     7.1295      -0.320993        24.6687   2.48485     2.94544     0.98336              41.961         13.5    61.35      5
   7 │ 5_13p5_61p35-6   5_13p5_61p35  22.2261     6.78449     -0.353836        24.5469   2.47104     2.94414     0.982008             40.0362        13.5    61.35      5
   8 │ 5_13p5_61p35-7   5_13p5_61p35  22.2123     6.55907      0.46061         24.4111   2.46496     2.94133     0.983553             36.1651        13.5    61.35      5
  ⋮  │        ⋮              ⋮            ⋮           ⋮            ⋮            ⋮            ⋮         ⋮           ⋮                 ⋮               ⋮        ⋮       ⋮
  31 │ 5_8p4_28p68-9    5_8p4_28p68   40.4129    15.381       35.4248          85.391    5.28316    14.6976      0.628366             54.0731         8.4    28.68      5
  32 │ 5_8p4_28p68-10   5_8p4_28p68   40.729     20.3822      33.0557         566.571    1.88684    26.3958      0.491463             54.3218         8.4    28.68      5
  33 │ 5_8p4_28p68-11   5_8p4_28p68   41.6894    18.5982      35.8152         707.16     1.50947    28.0554      0.509829             56.0121         8.4    28.68      5
  34 │ 5_8p4_28p68-12   5_8p4_28p68    0.700078  -0.0662316   -0.00804089      21.3812   2.64158     2.94514     0.96161               0.0593394      8.4    28.68      5
  35 │ 5_8p4_28p68-13   5_8p4_28p68   39.0276    18.1646      31.7101         660.916    1.50947    27.9988      0.504366             51.1676         8.4    28.68      5
  36 │ 5_8p4_28p68-14   5_8p4_28p68   12.5579    -0.761769     0.545299        15.3284   0.377368    2.89432     0.917512              1.12274        8.4    28.68      5
  37 │ 5_8p4_28p68-15   5_8p4_28p68   47.225     17.9276      40.9265         152.034    5.28316    19.533       0.609                62.9841         8.4    28.68      5
                                                                                                                                                           22 rows omitted

source

Filtering based on collapsed data

After collapsing the data, its common to filter out microbots that may be too small or large, going too slow, or stuck to the substrate. This can be performed using the filter_trajectories function.

MicroTracker.filter_trajectories — Function

filter_trajectories(collapsed_data::AbstractDataFrame, filter_settings::NamedTuple)

Filter out microbots that may be too small or large, going too slow, or stuck to the substrate.

The filter_settings is a NamedTuple with the following fields:

MIN_VELOCITY : minimum velocity in um/s
MIN_BOUNDING_RADIUS : minimum bounding radius in um
MAX_BOUNDING_RADIUS : maximum bounding radius in um
MIN_DISPLACEMENT : minimum total displacement in um

Example

filter_settings = (
    MIN_VELOCITY = 10.0,  # um / s  
    MIN_BOUNDING_RADIUS = 3.38,  # um
    MAX_BOUNDING_RADIUS = 75,  # µm
    MIN_DISPLACEMENT = 0,  # µm
)

filtered_collapsed_data = filter_trajectories(collapsed_data, filter_settings)

source