Neural connectivity

Wavelet analysis

As the volumes of data collected from neural tissue with optical imaging increases, and as the use of integrated optical systems becomes more ubiquitous, there is an added need for more sophisticated system analysis tools. To study the transient dynamics of cellular connectivity, we developed a time-varying connectivity algorithm for optical imaging data. The data in particular was acquired from neural cultures with GCaMP6s, and the connectivity between cells was inferred from their calcium dynamics. A time-varying window for the Pearson’s correlation coefficient was used to determine the linear connectivity between cells, and more concretely, the wavelet transform of the data was used to acquire the frequency-time information exchanged at different instances of cellular activity. Notably, wavelet coherence can be used to also identify the strength of neural connectivity at any point in time, the frequencies at which this is most prominent, and the phase information at these instances between cells, which provides relevant information about the directionality of signal exchange. Coupled to optical imaging, this technique provides a robust method for comprehensively characterizing the communication mechanisms from optical information exchange between cells.

  • Renteria C, Liu Y-Z, Chaney EJ, Barkalifa R, Sengupta P, Boppart SA. Dynamic tracking algorithm for time-varying neuronal network connectivity using wide-field optical image video sequences. Scientific Reports, 10:2540, doi:10.1038/s41598-020-59227-5, 2020.

Top: Connectivity workflow for calcium imaging videos, highlighting the different types of data acquired for connectivity analysis. Bottom: Representative calcium imaging plots of cellular activation for multiple cells in a culture (top), the overall connectivity weights acquired between cells (middle), and the time-varying connectivity plots between pairs of cells (bottom).