katzlabbrandeis

Boosting Ephys Data Analysis with the New Utility Addition

Visual representation of Add ephys_data as a utility

Date: August 29, 2024
Contributors: Abuzar Mahmood, abuzarmahmood
PR: https://github.com/katzlabbrandeis/blech_clust/pull/215

Introduction

In our continuous efforts to improve the Blech Clust project, we have made some significant changes that warrant your attention. This blog post will detail the recent addition of the ephys_data utility, a change aimed to enhance the project’s usability and functionality.

Key Technical Aspects

Primarily, the pull request incorporates the ephys_data utility into our codebase. This addition involved changes to 11 files, with 1939 lines added and no deletions. The impacted files are primarily in Python and Log languages.

The main files changed in this pull request include:

utils/ephys_data/BAKS.py
utils/ephys_data/convenience_scripts/region_spectrogram_plot.py
utils/ephys_data/ephys_data.py
utils/ephys_data/lfp_processing.py
utils/ephys_data/project.log

Among the changes, one crucial update was made to the ephys_data utility’s BAKS.py file. Here’s a snippet of the updated code:

"""
Python implementation of BAKS from 10.1371/journal.pone.0206794
"""
import numpy as np
import scipy.special 

def BAKS(SpikeTimes, Time):
    N = len(SpikeTimes)
    a = float(4)
    b = float(N**0.8)
    sumnum = float(0); sumdenum = float(0)
    
    for i in range(N):
        numerator = (((Time-SpikeTimes[i])**2)/2 + 1/b)**(-a)
        denumerator = (((Time-SpikeTimes[i])**2)/2 + 1/b)**(-a-0.5)
        sumnum = sumnum + numerator
        sumdenum = sumdenum + denumerator
    ...

Impact and Benefits

The addition of ephys_data as a utility brings a number of benefits. First, it improves the project’s usability by providing a defined way to handle electrophysiological data. It also aids in producing more accurate analysis results, thanks to the BAKS (Bayesian Adaptive Kernel Smoother) method implemented in the BAKS.py file for spike train data smoothing.

Moreover, the changes made in the region_spectrogram_plot.py file, which is part of the convenience_scripts, will allow users to generate and plot spectrograms for specific regions more conveniently.

Conclusion

In conclusion, this pull request represents a significant stride forward in our project’s development, making it more user-friendly and functional. We believe that the addition of the ephys_data utility will greatly enhance data analysis and visualization capabilities for all users. We’re excited for you to try out these new features and look forward to your feedback!