How to synchronize single-cell timeseries over a population

This guide shows you how to collapse single-cell signal traces into population-averaged time series, aligned to a reference event time.

Reference keys: mean signal, signal response, population average

Prerequisite: You have segmented, tracked, measured, and annotated events for a cell population.

Step 1: Configure the signal plot

Synchronized signals. The mean RICM intensity of the effector cells of the spreading demo, aligned on their spreading time.

1. Go to the Analyze tab and click plot signals (1).

2. In the Options window, configure the following:

   • population: The cell population (or pair) to analyze (2).

   • class: The column used to segregate cells (e.g., class_spread). This determines the “event” vs “no event” grouping.

   • time of interest: The event time column (e.g., t_spread) used to align the traces (t=0).

   • cmap: (Optional) Select a colormap for the curves.

   • absolute time: Check this to ignore the event time and synchronize signals using an absolute frame number (set via the slider) (3).

   • select cells with query: (Optional) Enter a pandas query to filter cells (e.g., TRACK_ID > 10) (4).

   • time calibration (frame to min): Frame-to-minute conversion factor.

   • pool projection: Choose how to aggregate the population (mean or median).

   • min # cells for pool: Minimum number of cells required to calculate a valid data point.

3. Click Submit (5).

Step 2: Select the signal

1. A second window appears (“Select numeric feature”).

2. Select the measurement you want to plot (e.g., adhesion_channel_mean) (6).

3. Click set (7).

Step 3: Interact with the plot

The plot window displays the synchronized signals. You can interact with it using the controls:

• Grouping (8): Switch between well (pooled per well), position (per position), or both.

• Toolbar Buttons (9):

  • Legend: Toggle legend visibility.

  • Log: Toggle log-scale for Y-axis.

  • CI: Toggle 95% confidence intervals.

  • Cell lines: Toggle display of individual single-cell traces.

  • Export: Save the figure or export tabular data.

• Class of interest (10): Filter the displayed curves by class:

  • *: Show all cells.

  • event: Show only cells belonging to the event class (class 0).

  • no event: Show only cells belonging to the non-event class (class 1).

• Rescale (11): Set a scaling factor for the Y-axis and press rescale.

• Single-cell signal alpha: Adjust the transparency of individual cell traces.

• Select position (12): Choose which positions/wells to display, either by name (checkboxes) or spatially (clicking on the position map).