Mini Analysis Plugin

The purpose of this plugin is to detect, analyze, and annotate mini synaptic events.

../_images/mini_analysis.png

A typical layout of the Mini Analysis Plugin.

Requirements

Tab title: Mini

Analysis Buttons

Find all

Automated analysis of the recording from start to finish. Only the current channel is considered.

Delete all

Delete all minis in the current channel.

Find in window

Automated analysis of the recording in the visible x-axis. Only the current channel is considered.

Delete in window

Delete minis annotated in the visible x-axis.

Report stats

Reports the summary of the discovered minis to the results-panel.

Open guide

Opens the mini guide popup window.

Core Parameters

Direction

Indicates the expected sign of the minis. Typically, the direction is positive for potential and negative for current recordings.

Search radius in % of the visible x-axis (Manual)

Used for manual mini analysis. Indicates the x-axis radius that is considered for analysis when the mouse is clicked. The radius is represented as a percentage of the total x-axis. In general, the lower number in this parameter, the higher the precision of the mouse-click.

Search window in ms (Auto)

Used for automated mini analysis. Indicates the window of x-axis that is considered at a time. The window traverses from left to right, sampling the specified window at a time. In general, the lower number in this parameter, the fewer minis are skipped but also the slower the analysis.

Caution

The x-axis window considered for each mini analysis must be larger than at least of 20 data points worth of x-axis range (i.e. for a 10kHz recording, the window must be larger than 2ms) Ideally, each mini analysis covers at least the approximate width of a single mini event.

Window before peak to estimate baseline (ms)

The baseline is calculated by averaging data points. Setting this value >0 fixes where the data points are sampled in relation to the peak time to calculate the baseline. The value specifies the left most (earliest) data point to sample for baseline calculation.

Set this parameter to 0 to ignore the parameter and use the trailing-moving-average.

Window averaged to find start of mini (ms)

The baseline is calculated by averaging data points. This parameter sets the size of the x-axis window that should be sampled to calculate the average. Higher numbers tend to result in increased robustness against consistent baseline noise; however, higher numbers can result in effects of unusually large and rapid deviations from the baseline (i.e. artefacts) affecting the surrounding mini baseline calculations.

Use decay to extrapolate halfwidth

If a mini does not decay below 50% of the amplitude, halfwidth cannot be calculated. Checking this option allows SimplyFire to estimate the halfwidth using the decay properties.

Minimum valley size in % of peak amplitude

Input available if analyzing compound minis. Enter a percentage of the maximum amplitude that must be reached during the decay phase before the next mini.

Ignore minis closer than (ms)

Input available if analyzing compound minis. Enter the radius around the peak that should be ignored while searching for compound minis. Increase this value if the automated analysis is picking up noise around the peak.

Decay calculation method

Two algorithms are available to estimate the decay constant (tau): Curve fit or % amplitude Curve fit uses the scipy package to fit a single exponential to the raw data. % amplitude uses the first data point that reaches a certain percentage of the maximum amplitude after the peak to estimate the decay constant.

Starting seed for exponential decay fit (ms)

Input available for Curve fit algorithm. Enter the ‘best guess’ for the decay constant when fitting a single exponential to the decay phase of the mini. This parameter is passed to the scipy.optimize.curve_fit function.

Percent peak to mark as decay constant (%)

Input available for % amplitude algorithm. Enter the percentage value used to estimate the decay constant. Enter 37 to estimate the decay constant for a single exponential function.

Maximum x-interval considered for decay (ms)

Input available for Curve fit and % amplitude algorithms. Enter the maximum range of data points that should be sampled to calculate the decay constant.

Analyze compound minis

Select this checkbox to analyze compound minis

Maximum interval between two peaks to use compound mini analysis (ms)

Input available if analyzing compound minis. Enter the maximum range between two minis that should use compound mini analysis.

Apply button

Apply the parameters for future analysis.

Default button

Sets the core input parameters to default values.

Filtering parameters

Minimum and maximum values can be set for the following parameters. Setting these values before analyzing minis will reject any candidate minis that do not meet the parameters. Previously found minis can be filtered by setting the parameters and clicking on Apply filter (all) or Apply filter (window).

Minimum values set to 0 and maximum values set to None will be ignored.

amplitude (absolute value)

Specify the minimum/maximum amplitude required for a mini. Use absolute value (unsigned). Setting this to 0 (without other filtering parameters) will result in most noise being annotated as a mini.

decay constant (tau) (ms)

Specify the minimum/maximum decay constant required for a mini in ms.

halfwidth (ms)

Specify the minimum/maximum halfwidth required for a mini in ms. Halfwidth is calculated as the minimum time between data points at 50% of the maximum amplitude during the rise and decay phases.

rise constant (ms)

Specify the minimum/maximum rise time required for a mini in ms. Rise is calculated from the start of the mini (0% amplitude) and the peak (100% amplitude).

decay/rise ratio

Specify the minimum/maximum ratio between decay and rise for a mini.

signal-to-noise ratio

Specify the minimum/maximum ratio between baseline noise standard deviation and the maximum amplitude or the mini. Baseline noise is defined as the standard deviation of the data points sampled in estimating the baseline.

Confirm button

Sets the filtering parameters for future analysis. This does not affect previously found minis.

Default

Sets the filtering parameters to default values. This does not affect previously found minis.

Apply filter (all) button

Applies the filtering parameters to all previously found minis. Any minis that do not meet the criteria are discarded.

Apply filter (window)

Applies the filtering parameters to previously found minis in the visible x-axis. Any minis within the visible x-axis that do not meet the criteria are discarded.

Data-panel display columns

Toggle the checkboxes to change the visibility of columns in the data-panel.

Show all button

Sets all columns in the data-panel to ‘show’.

Hide all button

Hides all columns in the data-panel.

Style setting

The markers on the plot for minis can be altered in the Style Plugin control-panel.

The size and color of the following markers can be changed. Size should be an integer. Color should be a hex code or a color name accepted by matplotlib. See the matplotlib documentation for more details on supported color names.

Peak

Marks the data point in the mini with the maximum amplitude as a dot.

Start

Marks the data point at the start of the mini with ‘x’.

Decay

Marks where t=decay constant with ‘x’. Depending on the decay algorithm chosen, the point may or may not be an actual data point in the recording.

Highlight

Marks only the selected minis with a dot. Minis can be selected by clicking on existing peak markers or by left-click + drag.

Apply button

Apply the stylistic parameters

Default button

Sets the input parameters to default values

Mini guide

Sometimes it can be unclear why a mini was chosen or why the parameters look different from what was expected. The mini guide attempts to clear such questions by providing a detailed report of a given mini.

Click the Open guide button to open a popup. Triggering a manual analysis or selecting a single mini will plot the same mini in the popup window.

../_images/mini_guide.png

An example mEJC shown in the mini guide

In the plot area, the following details are plotted:

  • Raw data points including and around the mini

  • The data point representing the start of mini

  • The data point representing the peak

  • Data points sampled to calculate the baseline

  • Single exponential decay function fitted to the decay phase

  • A point within the decay function where t = decay constant

  • A line plotting the baseline (a single y-value for a regular mini, extrapolated decay from previous mini for compound minis)

  • A line representing the amplitude

  • A line connecting the two data points used to calculate the halfwidth

In the text box below the plot, the following details can be found:

  • Whether or not there was a mini in the region

  • Peak data point

  • Baseline value

  • x-axis limits used to calculate the baseline

  • Amplitude

  • Rise constant (0-100)

  • Decay (tau)

  • Decay/rise ratio

  • Halfwidth

  • Signal-to-noise ratio

Tip

If no mini is discovered with manual analysis, the reason for the failure will be available in the text box. Adjusting the parameters or easing the filtering criteria may help detect the mini

Remove Restrictions button

Clicking this button when a manual analysis fails to detect a mini re-analyzes the region but without any of the filtering parameters.

Reanalyze button

Clicking this button when a manual analysis fails to detect a mini re-analyzes the region with the parameters set in the control-panel. Use this button to reanalyze minis after changing parameters.

Reject button

Clicking this button when a mini is highlighted discards the mini from the data.

Data-panel

Every mini marked in the trace plot is also annotated in the data-panel.

An entry can be selected by left click

Multiple entries can be selected by left click + Shift or left click + Ctrl. Selecting entries in the data-panel also highlights minis in the trace plot, adjusting the x- and y-axes so that all the selected minis are visible. Conversely, minis selected from the trace plot are also selected in the data-panel.

Pressing Delete/Backspace/e deletes the entries in the data-panel and associated mini markers on the trace.

Mini data are cleared when a new recording is opened.