First report

I’ve started the PhD on 1st October 2010. This is the first in the series of weekly reports, so it’s probably going to be a bit longer.

So far I have been looking into the literature on synaptic integration in general and more specifically on the influence of noise on the integration (find the list of papers I have been reading below).

As you can see from the list of papers, I was mainly focussing on the work of Destexhe with synaptic background activity as the source of noise. For this we planned to first computationally reproduce their results (responsiveness of pyramidal neurons in the presence of background activity is increased and becomes probabilistic [2,4,7]). Then we could use this model (or at least the lessons learned) to study the ability of neuron to differentiate synaptic input sequences and/or dependence of responses on the input location.

We had a Skype meeting (with Mark and Clemens) and discussed this idea, where Clemens gave us the overview on the current status on realistic neuron activity in leaving mammals, conflicting with Destexhe assumption of high-conductance state of neocortical neurons in vivo. This calls for rethinking and some more literature reading.

I also started using neuron and Morphforge, which is a set of convenience interface scripts developed by Michael Hull. This seems like a great tool for further modelling work.

1. van Rossum M, O’Brien BJ, Smith RG. Effects of Noise on the Spike Timing Precision of Retinal Ganglion Cells. J Neurophysiol. 2003;89(5):2406-2419.
2. Destexhe A, Rudolph M, Pare D. The high-conductance state of neocortical neurons in vivo. Nat Rev Neurosci. 2003;4(9):739-751.
3. Faisal AA, Selen LPJ, Wolpert DM. Noise in the nervous system. Nat Rev Neurosci. 2008;9(4):292-303.
4. Destexhe A, Pare D. Impact of Network Activity on the Integrative Properties of Neocortical Pyramidal Neurons In Vivo. J Neurophysiol. 1999;81(4):1531-1547.
5. Jia H, Rochefort NL, Chen X, Konnerth A. Dendritic organization of sensory input to cortical neurons in vivo. Nature. 2010;464(7293):1307-1312.
6. Branco T, Clark BA, Häusser M. Dendritic discrimination of temporal input sequences in cortical neurons. Science. 2010;329(5999):1671-1675.
7. Ho N, Destexhe A. Synaptic Background Activity Enhances the Responsiveness of Neocortical Pyramidal Neurons. J Neurophysiol. 2000;84(3):1488-1496.
8. Boucsein C, Nawrot M, Rotter S, Aertsen A, Heck D. Controlling Synaptic Input Patterns In Vitro by Dynamic Photo Stimulation. J Neurophysiol. 2005;94(4):2948-2958.

This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to First report

  1. mark says:

    Hi MIha

    It would be a good idea to start thinking about correlations. After all we expect neural input correlated. Both between excitatory inputs and across inh-exc. one expects correlations (Okun & Lampl paper, also check w. Paolo).
    Destexhe has a particular way to do this (you can also check the methods in my 2000 STDP paper), but there are many ways. In particular, one would expect a smeared temporal correlation, not a delta function. In addition, higher order correlation might be present.

    To get started with the simulations it would be good to simply reproduce the Ho and Destexhe data.

    PS. Did you receive the preprint of the review by Clemens?

    • Miha says:

      Hi Mark,

      thanks for comments and ideas. I’ll look into the correlations. I liked the way Destexhe achieves it (having a way to “tune” the correlation coefficient).

      I haven’t received the preprint from Clemens yet. Last week he told me the current version is with Ad Aertsen now and he would send it to me after he gets it back from him. He estimated that would happen this week. I’ll send him a reminder by the end of the week.

      I hope you will receive this reply automatically, I am still fiddling a bit with the system, so let me know if there is something not working out.

    • mark says:

      Concerning the correlations:
      Yes, Destexhe method is neat, but it is not the only way.
      I suspect that this might be an under-constrained problem:
      given observed pairwise correlations (and their temporal shape), there might be a whole family of process that lead to that 2nd order correlation, but vary in higher orders.

      Therefor it is good to understand this.

  2. mark says:

    Note, the Scanziani talks at the Santa Barbara KITP website is of interest.