Skip to main content

Advertisement

Extracting activity of individual cell populations from multielectrode recordings

Article metrics

  • 823 Accesses

  • 2 Citations

The low-frequency part of the extracellular electric signals, the local field potentials (LFP), carries information about dendritic processing in neuronal populations. However, the long-range nature of electric field makes the analysis of LFP difficult, as typically an electrode records activity of many sources. Modern multielectrodes allow for increased spatial resolution, hence also the need for effective data analysis methods which would allow to get more insight. These methods include Current Source Density (CSD) analysis and source separation methods.

In [1] we have combined inverse CSD method [2] with Independent Component Analysis (ICA) [3] to decompose activity recorded in the rat forebrain on a grid of 140 positions obtaining physiologically plausible components across a group of seven animals. The question remains how the obtained components are connected to the activity of neuronal populations. To study this problem we enriched the thalamocortical model [4, 5] by adding the spatial information (Figure 1A) and used it to simulate the LFP generated by a single cortical column. We used the kernel CSD method and spatio-temporal ICA to decompose the LFP measured on a regular grid. We compared the resulting components to the activity of the twelve cortical populations included in the model. We found that the recorded evoked activity was dominated by two populations of pyramidal neurons, which were well separated by ICA (Figure 1B-1G). The activity from other populations was hardly visible on top of the main two dipoles and we were also not able to obtain them through ICA.

Figure 1
figure1

A) Model setup. B) Reconstructed CSD generated in a thalamo-cortical column 20 ms from the onset of stimulation. C) Activity of superior pyramidal cells (rhythmic spiking) in layer 2/3. D) Activity of tufted pyramids (intrinsically bursting) from layer 5. E) sum of two most prominent ICA components shown in F) and G) separately.

References

  1. 1.

    Łęski S, Kublik E, Świejkowski DA, Wróbel A, Wójcik DK: Extracting functional components of neural dynamics with ICA and iCSD. J Comput Neurosci. 2010, 29: 459-473. 10.1007/s10827-009-0203-1.

  2. 2.

    Łęęski S, Wójcik DK, Tereszczuk J, Świejkowski DA, Kublik E, Wróbel A: Inverse Current-Source Density in three dimensions. Neuroinformatics. 2007, 5: 207-222. 10.1007/s12021-007-9000-z.

  3. 3.

    Stone JV, Porrill J, Porter NR, Wilkinson IW: Spatiotemporal Independent Component Analysis of Event-Related fMRI Data Using Skewed Probability Density Functions. NeuroImage. 2002, 15: 407-421. 10.1006/nimg.2001.0986.

  4. 4.

    Traub RD, Contreras D, Cunningham MO, Murray H, LeBeau FEN, Roopun A, Bibbig A, Wilent WB, Higley MJ, Whittington MA: Single-Column Thalamocortical Network Model Exhibiting Gamma Oscillations, Sleep Spindles, and Epileptogenic Bursts,. J Neurophysiol. 2005, 93: 2194-2232. 10.1152/jn.00983.2004.

  5. 5.

    [http://senselab.med.yale.edu/ModelDb/ShowModel.asp?model=82894]

Download references

Acknowledgements

Supported by grants PBZ/MNiSW/07/2006/11, 5428/B/P01/2010/39 from the Polish Ministry of Science and Higher Education and by a grant from the Polish Ministry of Regional Development, POIG.02.03.00-00-003/09.

Author information

Correspondence to Daniel K Wójcik.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Potworowski, J., Głąbska, H., Łęski, S. et al. Extracting activity of individual cell populations from multielectrode recordings. BMC Neurosci 12, P374 (2011) doi:10.1186/1471-2202-12-S1-P374

Download citation

Keywords

  • Independent Component Analysis
  • Neuronal Population
  • Source Separation
  • Local Field Potential
  • Current Source Density