Template Metrics

This extension computes commonly used template metrics. A template is the average (or median) spike waveform for each unit.

First, the templates are upsampled (optionally, default 10) to increase the temporal resolution for more precise metric computation. Then they are (optionally) smoothed (using a Savitzky-Golay filter) to reduce high-frequency noise that could interfere with peak/trough detection.

We compute the channel where the extremal value of your template occurs. Many single channel metrics only depend on the template on this channel, while multi-channel metrics will depend on the templates from many channels of your probe. On the extremal channel, we compute the troughs (minima) and peaks (maxima) of the template.

Peaks are divided into “peaks before trough” and “peaks after trough”, depending on their temporal location with respect to the main trough.

Visualization of trough, peaks, and a few metrics.

Here is a list of metrics computed using the template on the main extremum channel:

half_width

Duration in seconds at half the amplitude of the largest trough (minimum) and peak of the template. It generates two columns: trough_half_width and peak_half_width.

peak_to_trough_duration

Duration in seconds between the main trough and the main peak after the trough.

main_to_next_extremum_duration

Duration in seconds from the main extremum (peak or trough) to the next (in time) extremum. If no next extremum is found, it returns NaN.

repolarization_slope

Slope of the repolarization phase of the template, between the trough (minimum) and return to baseline in µV/s. If there is no trough, it returns NaN. If the absolute maximum is a peak, it still uses largest trough to compute.

recovery_slope

Slope of the recovery phase of the template, from the first peak following the largest trough, returning to baseline in µV/s. If there is no trough, it returns NaN. If there is no peak after the main trough, it returns NaN.

Note

The units of recovery_slope and repolarization_slope depend on the input. Voltages are based on the units of the template. By default this is µV but can be the raw output from the recording device (this depends on the return_in_uV parameter, read more here: SortingAnalyzer). Distances are in µm and times are in seconds. So, for example, if the templates are in units of µV then recovery_slope repolarization_slope is in µV/s.

waveform_ratios

This metric includes several ratios between peaks and troughs:

main_peak_to_trough_ratio

Ratio between the amplitude of the main peak (before or after trough) and the absolute amplitude of the largest trough. If no troughs, or no peaks are found, it returns NaN.

peak_before_to_trough_ratio

Ratio between the amplitude of the peak before the main trough and the absolute amplitude of the largest trough. If no troughs, or no peaks before trough are found, it returns NaN.

peak_after_to_trough_ratio

Ratio between the amplitude of the peak after the main trough and the absolute amplitude of the largest trough. If no troughs, or no peaks after trough are found, it returns NaN.

peak_before_to_peak_after_ratio

Ratio between the amplitude of the peak before the main trough and the amplitude of the peak after the main trough. If no troughs, or no peaks before trough or no peaks after trough, it returns NaN.

waveform_widths

This metric includes the widths of peaks and troughs, given by the scipy.signal.find_peaks function.

trough_width

Width in seconds of largest trough.

peak_before_width

Width in seconds of peak before largest trough.

peak_after_width

Width in seconds of peak after largest trough.

number_of_peaks

This metric includes the number of positive and negative peaks in the waveform: num_positive_peaks and num_negative_peaks.

For recordings with a large number of channels, it can be useful to compute multi-channel metrics that take into account the spatial distribution of the template across the probe (these are computed by default if the number of channels is greater than 64, but can be forced on or off with the include_multi_channel_metrics parameter).

These are the multi-channel metrics that can be computed:

velocity_fits

Computes the velocity of the template propagation along the probe using a linear fit of the time delays of the troughs across channels versus the distance from the channel with the largest amplitude. It returns two columns: velocity_above and velocity_below, which are the velocities in µm/ms above and below the max channel of the template, respectively.

exp_decay

Exponential decay in 1/µm of the template amplitude over distance from the channel with the largest amplitude. Optionally, the fit can also be linear (default is exponential).

spread

Spread in µm of the template amplitudes, defined as the distance between the channels “observing” a template above 20% of the maximum amplitude (default). Template amplitudes are normalized to the maximum amplitude and optionally smoothed over space using a Gaussian filter (default sigma is 20µm).

tm = sorting_analyzer.compute(input="template_metrics", include_multi_channel_metrics=True)

For more information, see compute_template_metrics()