to Table of Contents

4. Event Detection

Event detection function detects events according up to four conditions in time course plots either of A/D sweep ("Event detection..." menu item in the "A/D option" menu) or time lapse plots of image movie ("Event detection..." menu item in the pop up menu shown by right-button clicking (or clicking with control key pressing) in a plot window). Results are displayed in the time course plots (or can be turned off), and can be sent to Igor as histograms. Detected events can be further analyzed by the Event Ensemble Analysis function.

4.1 Event Detection Panel

• channel

  Searched channel (1..); displayed in the case for A/D data
  

• search data range

  Range of data points for event detection.

  

  • all data

    Search range is whole data (not effective in on-line)

  • displayed range

    Search range is limited to the currently displayed range of the time course plot (not effective in on-line detection)

• base line
  • show baseline

    baseline is displayed on the time course plot (dotted gray line in the figure).

  • detection modes
  • fixed value
    A fixed value is used as a baseline value.
  • linear regression
    Baseline is calculated by extrapolating linear regression of values within the "detect window" time range just before each point.
    
  • average
    Baseline is calculated by averaging values within the "detect window" time range just before each point.
    
  • baloon
    experimental.
    
  • asymmetric least sq
    Asymmetric Least Squares Smoothing baseline correction algorithm (P. Eilers, H. Boelens 2005). iteration, weight and smooth factor should adjusted.
  • asym reweight penalized least sq
    Asymmetrically reweighted penalized least squares smoothing (Sung-June Baek, Aaron Park, Young-Jin Ahna and Jaebum Choo, Analyst, 2015, 140, 250 (2015)). iteration, weight and smooth factor should adjusted. This method is used in Ca event detection in Adachi et al. iScience 25:105138 (2022, doi:10.1016/ j.isci.2022.105138).
• display mode
  • peak

    Detected peak points are displayed on time course plots (see the figure). 

  • range

    Range of detected events are displayed on time course plots (see the figure).

  • frequency

    Display frequency of events by bars with time course plots. 

• event detection parameters
• show thres
  displays event-detection threshold on the time course plot (colored dot line in the figure).
• M x s.d.
  Threshold for event detection is determined as standard deviation of the entire data excluding the detected event periods multiplied by the M number. If events  downward deflect, use minus value (as in the figure).
  
• Schmitt trigger
  If the "Schmitt trigger" is checked, beginning of each event is detected as explained, but the event is terminated when a data amplitude does not exceed the "post threshold". In the Schmitt trigger mode, the "permissible gap width" is not taken into account.
• minimum time window
  Minimum duration of an event (events of shorter duration than this are ignored).
• permissible gap width
  If it is more than zero, gap periods (period for data amplitudes are not exceeding the threshold) less than the "permissible gap width" are counted as a part of the event (see the illustration in the figure).
  

• slope detect window width
  Slope is calculated with this window (see the above scheme).
• peak average width
  Peak amplitude is calculated by averaging data points within a time window determined by the "pre" and "post peak average width" parameters.  If zero values are set to these parameters, amplitude of the peak point is reported.
  
• number of detected events.
• Analyze button 

  Executes event detection.

• Exit button 

  Exit this dialog with parameters kept.

• Cancel button 

  Exit this dialog with parameters not changed.

• set as default button

  Set the current parameters as a default set. 

• use default button 

  Use default parameter set.

4.2 Event Detection Information panel

"Event" panel appears after event detection is performed.

• Recalc: peak detection is executed by showing the "event detect panel".
  
• export
• ensemble: seel below ("Event Ensemble Analysis")
  
• to Igor
  Sends results to Igor as histograms.
• bin / num of bin ratio boxes
  Select bin condition from two ways. bin: set bin width manually, num of bin: set number of bins.
• whole histogram

  Makes a histogram of total data values.

• peak value

  Makes a histogram of peak values of detected points. 

• base line

  Makes a histogram of base line values of detected points.

• detected time (open time)

  Makes a histogram of duration of detected points (equivalent of open time in single channel analysis). 

• non-detected time (close time)

  Makes a histogram of duration between events (equivalent of close time in single channel analysis).

• slope

  Makes a histogram of slope values.

• event count

  Makes a event frequency graph.

• raise histogram

  Raise a graph window of histogram. If it is off, just data is sent to Igor but no graph is made in it.

  • Gaussian fit

    Histograms are fitted by Gaussian distribution, and fit result is also displayed. Event count graph is not fitted by Gaussian distribution.

4.3 Event Ensemble Analysis

Events detected in the Event Detection function can be further analyzed by this. Events in different recorded data can be joined by dragging the time course plot window or the top icon of A/D data window onto the bottom part of this panel.