FiberPhotometry

Auditable fiber-photometry analysis from acquired signals to inference across animals.

FiberPhotometry is a Python library and command-line workflow for scientists who need to understand how preprocessing choices affect their result—not simply obtain one corrected trace. It preserves subjects, sessions, events, channels, exclusions, parameters, and uncertainty throughout the analysis.

Development status

This is a pre-release research tool. The documentation distinguishes supported, experimental, planned, and out-of-scope methods. Experimental availability is not a claim of scientific validation.

Find your question

I want to…	Start with
Run an event-aligned comparison across animals	First event analysis
Analyze CSV/TSV exports without rewriting code	Configuration-first CLI
Import a TDT block	TDT import
Work from public NWB data	DANDI tutorial
Combine DeepLabCut, SLEAP, MoSeq or BORIS with photometry	Behavioral ecosystem tutorial
Separate calibrated optical contributions	Wavelength-aware optical unmixing
Test or invert a declared sensor response	Sensor-kinetic modeling
Analyze a coordinate-mapped multi-fiber array	Coordinate-aware dense arrays
Model neural summaries across learning	Unspool interoperability
Describe a long recording at several time scales	Multiscale long-duration summaries
Compare reasonable preprocessing choices	Robustness multiverses
Choose an inferential method	Methods catalog
See what the package cannot yet do	Capability matrix
Produce verifiable publication evidence	Publication workflow

The evidence path

The evidence path. Signal identity, clocks, events, processing choices, exclusions, uncertainty, and provenance remain attached to the final scientific claim.

The project does not claim that one preprocessing or statistical method is always correct. It makes the choice, its assumptions, and its sensitivity visible.