Fitting neuron models to spike trains

Frontiers in Neuroscience

Dan F. M Goodman¹  Bertrand eFontaine¹  Romain eBrette¹  Cyrille eRossant¹  Anna K Magnusson⁴  Jonathan ePlatkiewicz⁵ 

[1] CNRS;Ecole Normale Supérieure;Karolinska Institutet;Karolinska University Hospital;UPMC Univ Paris 6;

关键词: simulation;    optimization;    brian;    gpu;    model fitting;    Parallel Computing;   

DOI  :  10.3389/fnins.2011.00009

来源: DOAJ

【 摘 要 】

Computational modeling is increasingly used to understand the function of neural circuitsin systems neuroscience.These studies require models of individual neurons with realisticinput-output properties.Recently, it was found that spiking models can accurately predict theprecisely timed spike trains produced by cortical neurons in response tosomatically injected currents,if properly fitted. This requires fitting techniques that are efficientand flexible enough to easily test different candidate models.We present a generic solution, based on the Brian simulator(a neural network simulator in Python), which allowsthe user to define and fit arbitrary neuron models to electrophysiological recordings.It relies on vectorization and parallel computing techniques toachieve efficiency.We demonstrate its use on neural recordings in the barrel cortex andin the auditory brainstem, and confirm that simple adaptive spiking modelscan accurately predict the response of cortical neurons. Finally, we show how a complexmulticompartmental model can be reduced to a simple effective spiking model.

【 授权许可】

Unknown