Skip to content

Reads data from csv files into Python and bins data according to a user input threshold. Can graph the performance of different thresholds (against a curated dataset).

Notifications You must be signed in to change notification settings

nimra-asi/fluorescent-trace-analysis-in-Python

Repository files navigation

fluorescent-trace-analysis-in-Python

This is a collection of Python functions that do the following for fluorescent trace (imaging data) output from Fiji:

The traces (stored in csv files) are read into Python. The following processing is performed:

  1. bg subtraction
  2. each trace (representing a single roi or region of interest) is examined to pick out "peaks" (spontaneous exocytosis events) represented by a sudden increase in flourescence. This involves implementation of a threshold
  3. the performance of different thresholds can be tested automatically (false positives and false negatives against a curated set of data)
  4. once a threshold has been picked the events for different conditions can be averaged and graphed
  5. the sorted traces can be written to Excel

Because this is a collection of functions that together complete the goal task I've created a mock "run_client" file that demonstrates the order that the functions should be run as well as the Python libraries used

About

Reads data from csv files into Python and bins data according to a user input threshold. Can graph the performance of different thresholds (against a curated dataset).

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published