clampsuite 0.0.2

Slice electrophysiology analysis package for analyzing mEPSCs, o/eEPSCs and standard current clamp data.

ClampSuite

    ClampSuite is a suite of programs for analyzing slice electrophysiology data. The program can analyze data from ScanImage files (.mat) for current clamp, s/mEPSC, and o/eEPSC experiments. The program runs on both PC and Mac (it should run on Linux but it has not been test). ClampSuite should run on both newer and older computers (It runs perfectly well on my 7 year old MacBook Pro and iMac). In addition to providing a GUI for analysis you can call the acquisition classes used by the program to load your data into a Python script which is convient for creating figures for publication or for further analyzing your data. The program is not currently able to analyze other common electrophysiology file types however if you would like to use the program and have a file type that is not supported please send me some files and I can probably add support within a couple weeks.

    There are currently four different modules: MiniAnalysis, Current Clamp, oEPSC/LFP and Filter design. The program exports user settings for the interface, an individual JSON file for each acquisition, and an Excel file for the raw and processed data. Each module is built to allow the user to delete acquisitions, events or modify baselines or peak values. Each module allows for drag and drop to loading of files for analysis or to reload already analyzed files.

---

Modules

MiniAnalysis

    Minianalysis is setup to analyze both inward and outward currents using a variation of the fft deconvolution method as well as the original method fft deconvolution method. While the fft deconvolution requires a template, the template shape is not as crucial as it is for template matching algorithms. The module allows for control over filtering, offering a variatey for minimal-(Bessel, Butterworth), zero-phase filters (windowed FIR and Remez), median filter, and Savitsky-Golay filter.

    Currently the program analyzes amplitude of the event, rise time (time from baseline to the peak), rise rate (10-90%) and the estimated tau (2/3 peak amplitude or amplitude/e^1). The program can curve fit taus however, the curve fit function still needs some work and is mostly for visual inspection.

The program also shows the final data in a couple ways that I have found informative such as stemplot for all event parameters and a kde plot for cumulative events parameters.

oEPSC/LFP

     The oEPSC/LFP module analyzes optically/electrically evoked intracellular currents and/or LFP signals. For I/EPSCs the program finds the maximum current within a specified window. Windowed peak finding is necessary for NMDA current analysis if AMPAR . The program can also find the charge transfer, estimate the tau and curve-fit to find the tau. For LFPs the program finds the peak fiber volley, field potential, and the field potential rise slope. Future additions to the program will include multi-peak analysis.

Current Clamp

     The current clamp module analyzes evoked potentials. The program analyzes the delta-V, spike frequency, rheobase, and spike-threshold. Additional analysis on the first spike of acquisitions that contain spikes include, peak spike voltage, first spike peak velocity, peak afterhyperpolarization potential. The program can compile all the data for ease of analysis if the pulse amplitude is include in the analysis file. The program can analyze upwards of 400 acquisitions in around a minute.

Filter design

     The last module is a filter design tool. I initially built the GUI to experiment with different filtering techniques since filtering can have a huge impact on the quality of the analysis.

---

Output file types

    User settings files: The user settings are exported as a YAML file (essentially a text file) so that settings can be modified outside the program.

    Analyzed acquistion files: I chose to export acquisition data in a JSON file for a couple reasons. While JSON is not the most efficient for large arrays of data they are portable and can be opened in any programming language as well as a web browser. JSON files are also good for storing many different data types which are generated by the program. The reason I chose to generate a JSON file for each aquisition is that you can have all the data needed without opening a large file. The JSON file in Python just returns are dictionary which are easy to work with.

    Final data files: I chose excel files to output the raw and processed final data because they allow for bundling multiple dataframes into one file (one sheet per dataframe) and most people are familiar with this filetype.

---

To Do

• Add multi-peak finding for o/eEPSC module.
• Add acquisition manager to make it easier to integrate other file types and add load data for publication or further analysis
• Improve mini analysis decay curve fit.
• Save last used user settings.
• Add more color themes.
• Add performance stats to monitor memory and CPU usage.
• Integrate logging for debugging and performance issues.