Knowledge Base

Analyze Data

7.1 Select the Data

Using a Data Segmentation profile, you can select one part from each classified run that complies with three criteria (the minimum number of consecutive steps with a specific speed range and maximum allowed speed variation). This is useful, for example, if you have a run in which the animal briefly stopped at the beginning of a run but then continued to walk normally. You can automatically select only the 'good' part of the run for analysis and export with a Data Segmentation profile.

Default Data Segmentation Profile

The default Data Segmentation profile is labeled All Data. It cannot be deleted.

Procedure

1. Do one of the following:
   • In the Profiles Explorer, right-click Data Segmentation Profiles and select Add New Profile.
   • Choose Profiles > Data Segmentation Profile > New.
2. Enter the name of the new profile and click OK.
3. In the Profile window, under Select Data, you can set the following:
   • Minimum number of consecutive steps per run – Only the part of the run with the minimum number of consecutive steps which complies with the next two parameters is selected. We advise you to use at least 10 consecutive steps.
   • Average speed should range from (in distance unit/second) – Only the part of the run within the set speed range is selected. The Average Speed is based on the Body Speed of selected steps in a run. The Body Speed for each step cycle is calculated by dividing the distance that the animal's body has traveled from one initial contact to the next by the time it took to travel that distance. The Average Speed is different from the Run Average Speed (shown in the Acquisition and Classification screens), which is the average speed of the animal's body during the whole recorded run.
   • Maximum allowed speed variation (in percent) – Only the steps in a run within the set maximum allowed speed variation are selected. Example: The Average Speed of a run is 22.0 cm/s. The speed of a step cycle of the left front paw in this run is 29.07 cm/sec. The variation then is: ([29.07 - 22.0] / 22.0) * 100% = 32.1%.
   • Exclude non-compliant runs – When you select this option, runs not compliant with the run criteria in the Experiment Settings are excluded from the data selection.
   • Exclude partially classified runs – When you select this option, runs which have not been completely classified are excluded from the data selection.

Selected Data Overview Tab

Shows the result of your data segmentation. The sheet on the left shows, per treatment group and time point, the number of Selected Animals / Runs / Steps, the Total number of Animals / Runs / Steps in the experiment, and the Percentage of selected Animals / Runs / Steps. You can choose whether you want to see the selected Animals, Runs, or Steps by selecting the corresponding option at the bottom. The chart on the right is a graphical representation of the selected data. You can choose whether you want time points or treatment groups displayed on the X-axis by selecting the corresponding option at the bottom.

Complete Data Overview Tab

In the matrix on the left, the average speed intervals (1 to 10 cm/sec, 1 to 20 cm/sec, 1 to 30 cm/sec, etc., with centimeters as distance unit) are displayed on rows. The speed variation is displayed in 10% intervals on columns. Each cell in the matrix shows the resulting number of steps for a specific combination of Average Speed and Maximum Variation in speed based on the Minimum number of consecutive steps you selected.

In the Complete Data Overview, you can only change the Minimum number of consecutive steps per run. If you do this, you need to click the Calculate Complete Data Overview button at the bottom to re-calculate the number of selected steps.

The 3D chart shows:

• On the X-axis: the Average speed range intervals.
• On the Y-axis: the Maximum allowed speed variation intervals.
• On the Z-axis: the resulting number of steps per speed / speed variation interval.

What Speed or Variation Should I Choose?

• Minimum number of consecutive steps per run – We recommend selecting at least 10 consecutive steps. This generally corresponds to 1 step cycle for each paw (that is, each paw is placed twice on the glass plate). If you select a lower number, a paw might have been placed only once, which means that for this paw some parameters (for example, Swing or Stride Length) cannot be calculated.
• Average speed should range from ... to ... ([Distance unit]/second) – In the Complete Data Overview tab, the 3D chart shows the number of steps in the different Average Speed intervals; this helps you in determining the Average Speed range for your data. Select the widest average speed range possible for your runs if the Average Speed is different for different treatment groups (for example, the experimental animals are on average slower than the control animals). Steps with a maximum speed higher than 120 cm/sec (or an equivalent speed in a different distance unit) are not displayed in the 3D chart. However, you can select a maximum average speed of up to 300 cm/sec.
• Maximum allowed speed variation (percent) – By selecting a low or high speed variation, you allow respectively small and large differences in average speed for a run. Example: You have a run with an Average Speed of 25 cm/sec. You have set the Maximum allowed speed variation to 20 percent. The variation is calculated as follows: ([Step speed - Average Speed] / Average Speed) * 100 = speed variation (%). With these values, the step speed can range between 20 and 30 cm/sec. If you set the Maximum allowed speed variation to 60 percent, the step speed can range between 10 and 40 cm/sec.

Check the Data After Segmentation

While data segmentation is a powerful tool to remove outliers and generate more homogeneous samples which facilitate analysis, it is not free of risks. A subject under study may make a small jump that is typical of the studied phenotype and should remain in the data; however, data segmentation may remove a significant part of interesting data based on increased variation in speed or the number of consecutive regular steps. Always check the data resulting from data segmentation before running the analysis.

7.2 Select the Analysis Parameters

You can select specific parameters with Parameter profiles. Only selected parameters from the Paw Statistics and Phase Lags are displayed in the Run Statistics and in the Group Statistics and Charts, and exported to an Excel file. The default Parameter profile is labeled All Parameters and cannot be edited or deleted.

1. To create a new Parameter profile, do one of the following:
   • In the Profiles Explorer, right-click Parameter Profiles and select Add New Profile.
   • Choose Profiles > Parameter Profile > New.
2. Type in the name of the Parameter profile and click OK.
3. In the Parameter profile window, you can select or de-select the following:
   • Paws – Select the paws for which you want to display or export the Paw Statistics Parameters.
   • Parameters – Select the Paw Statistics Parameters you want to display or export.
   • Phase Dispersions – Select the paw combinations (Diagonal, Girdle, Ipsilateral) for which you want to calculate Phase Dispersions.
   • Couplings – Select the paw combinations (Diagonal, Girdle, Ipsilateral) for which you want to calculate Couplings.
4. Click OK to make the new Parameter profile active.

7.3 Visualize the Runs

Procedure: Open the Visualization Screen

• In the Experiment Tasks window, select View Run Visualizations.
• Choose Analyze > View Run Visualizations.
• In the Experiment Explorer, right-click a trial and select View Run Visualizations (only available when the trial contains one or more partially classified runs), or right-click a run and select View Run Visualizations (only available when the run has been partially or fully classified).

You can export the Visualizations. See Export the Run Visualizations for details.

Visualize a Run

1. Under Trial & Run Selection, select a trial and one of the runs in that trial, or click the Previous/Next button to navigate between trials and runs.
2. Replay the run as described in Review the Runs During Classification.
3. Under Visualizations, you can view:
   • 3D Footprint Intensities
   • 2D Footprint Intensities
   • Footfall Patterns

Paws which have not been selected in the currently active Data Segmentation profile are displayed in gray in the Video window and are not displayed in any of the Visualizations (3D/2D Footprint Intensities, Footfall Patterns). You can select another Data Segmentation profile by right-clicking it and selecting Set as Active.

3D Footprint Intensities

Click on the 3D Footprint Intensities tab. The 3D Footprint Intensities tab plots the print intensities of the 4 paws for each individual video frame in which the paw has contact with the glass plate in a 3D chart. The intensities range from 0 to 255. The color range used in the 3D chart is the same as is used in the False Color mode of the footprints in the Classification print view.

A 3D chart can be rotated in all directions: click a chart, keep the mouse button pressed, and drag to rotate the chart. To reset all charts to the original orientation, open another item in the experiment (for example the Trial List) and re-open the Visualization.