PASTa: An Open‐Source Analysis and Signal Processing Toolbox for Fiber Photometry Data

Figures

• 

  Figure 1. Overview of the PASTa protocol. PASTa includes an integrated pipeline for data preparation, signal processing, and transient detection and quantification.

• 

  Figure 2. First 30 s of a fiber photometry recording session [ventral tegmental area (VTA) dopamine (DA) activity, GCaMP6f], showing (A) the raw signal from the 465‐nm channel) and (B) the raw background from the 405‐nm channel. Both panels illustrate the initial large artifact immediately following the start of the recording. To mitigate undesired effects on signal processing, PASTa by default extracts the data and trims the first 5 s of each session.

• 

  Figure 3. Fiber photometry (VTA DA activity, GCaMP6f) raw (A) signal (465‐nm) and (B) background (405‐nm) data streams each fit with an exponential decay function (red). Both streams show a steeper decay slope in the first 2 min, reflecting an initially faster rate of photobleaching that stabilizes over time. To account for this, the PASTa protocol recommends cropping the data streams to exclude the first 2 min of each session prior to further analysis.

• 

  Figure 4. Representative example of default signal processing steps for a single recording session (VTA DA activity, GCaMP6f). (A) Raw signal (465‐nm) and raw background (405‐nm) streams. (B) FFT power spectrum of the raw signal and raw background streams with the power across frequencies from 0 to 100 Hz plotted on logarithmic scale. The scaling factor constant is determined in the frequency domain as the ratio of the power of the signal stream to the power of the background stream. By default, the power in all frequencies between 10 and 100 Hz (gray bar) is used to determine the scaling factor by which to multiply the background stream. (C) Raw signal (465 nm) overlaid with scaled background (scaled 405 nm). The background stream is multiplied by the scaling factor in the time domain and centered around the mean of the raw signal. (D) Power spectrum of the FFT of the raw signal and scaled background streams. (E) Subtracted signal (raw signal – scaled background) output as ΔF/F. (F) Zoomed‐in example of overlaid subtracted signal and subtracted and filtered signal. Filtering reduces high‐frequency fluctuations without significantly altering the shape of the signal. (G) Subtracted and…

• 

  Figure 5. Example output of the plotTraces function, which generates a tiled layout with the following fiber photometry data streams for a single recording session (VTA DA activity, GCaMP6f): (A) raw signal (465 nm), (B) raw background (405 nm), (C) raw signal overlaid with scaled background, (D) subtracted signal, and (E) subtracted and filtered signal.

• 

  Figure 6. Example output of the frequency‐domain representation of fiber photometry streams for a single recording session (VTA DA activity, GCaMP6f) created by the functions plotFFTpower and plotFFTmag. Top to bottom: FFT power spectral density (left) and magnitude (right) plots show the frequency content on a log scale for the following: (A) raw signal (465‐nm channel), (B) raw background (405‐nm channel; control), (C) overlaid raw signal and raw background, (D) overlaid raw signal with scaled background, (E) background subtracted signal, and (F) background subtracted and filtered signal. These plots allow visualization of how pre‐processing steps reduce high‐frequency noise while preserving low‐frequency neural signal components.

• 

  Figure 7. Normalized trace plot of a single recording session of VTA DA activity (GCaMP6f). Subtracted and filtered data are Z‐scored to the whole‐session mean and standard deviation (SD).

• 

  Figure 8. Comparison of transient detection with amplitude thresholds of 2, 2.6, and 3 SDs from a single VTA DA activity GCaMP6f recording session. The subtracted and filtered data stream was Z‐scored to the whole‐session mean and SD prior to transient detection. Note that each lower threshold includes the events detected by the higher thresholds (e.g., 2 SDs includes the events detected by both 2.6‐ and 3‐SD thresholds). (A) Normalized trace plot showing detected transients for each threshold. (B) Whole‐session transient count at each threshold. (C) Mean transient amplitude (Z‐score) at each threshold.

• 

  Figure 9. Transient detection and quantification. All peaks (dark blue) in the data stream are identified and compared to baseline (light blue), which can be determined from the pre‐peak window mean (default), pre‐peak window minimum, or pre‐peak local minimum. Amplitude (green) is calculated as the difference between the maximum peak value and the baseline, and events with amplitudes greater than the set threshold (red) are included. Transient events are quantified by amplitude (green), rise duration from half‐height (pink), fall duration from half‐height (purple), and half‐height area under the curve (AUC; yellow).

• 

  Figure 10. Example outputs from transient plot functions for a single fiber photometry recording session (VTA DA activity, GCaMP6f). The subtracted and filtered signal was Z‐scored to the whole‐session mean and SD prior to transient detection. The threshold for transient event detection was set at 2.6 SDs. (A) Whole recording session trace (green) with identified transient events (blue circles) generated by the function plotTransients. (B) Recording session trace divided into five 2‐min bins with detected transients (blue circles), generated by the function plotTransientBins. (C) Overlaid traces of all individual transients in the session aligned to peak maxima, generated by the function plotTransientTraces. (D) Overlaid aligned transient traces grouped by session bin, generated by the function plotTransientBinTraces.

• 

  Figure 11. Example analysis of VTA DA transient frequency and amplitude after acute morphine administration. Representative examples of pre‐injection baseline normalized traces from a single subject before and after (A) an injection of saline (control) and (B) an injection of morphine (10 mg/kg). The vertical dashed line indicates the time of injection. (C) Mean transient frequency (transients per minute) across the entire session. (D) Transient frequency binned in 5‐min time windows. (E) Mean transient amplitude (Z‐score) across the session. (F) Transient amplitude binned in 5‐min time windows. Data reflect automated transient detection using PASTa protocol defaults.

• 

  Figure 12. Example analysis of VTA DA transient event duration and AUC following acute saline (blue, control) and morphine (red) administration. (A) Mean transient rise duration (milliseconds) across the full session. (B) Transient rise duration binned in 5‐min time windows. (C) Mean transient fall duration (milliseconds) across the full session. (D) Transient fall duration binned in 5‐min time windows. (E) Mean AUC for transients. (F) Transient AUC binned in 5‐min time windows. All duration and AUC metrics were calculated at half‐height relative to the peak. The vertical dashed lines in (B), (D), and (F) indicate the time of saline or morphine injection.