# LFPy: A tool for simulation of extracellular potentials

Henrik Linden (Mathematical Sciences and Technology, Norwegian University of Life Sciences), Espen Hagen (Mathematical Sciences and Technology, Norwegian University of Life Sciences), Szymon Leski (Nencki Institute of Experimental Biology, Warsaw, Poland), Eivind S Norheim (Mathematical Sciences and Technology, Norwegian University of Life Sciences), Klas H Pettersen (Mathematical Sciences and Technology, Norwegian University of Life Sciences), Gaute T Einevoll (Mathematical Sciences and Technology, Norwegian University of Life Sciences)

Measurements of extracellular electrical potentials are one of the most important and common tools for probing neuronal activity. Extracellular electrodes are routinely used to count action potentials from neurons nearby by exploiting information contained in the high-frequency band of the signal (>~500 Hz). The low-frequency band (<~300Hz), termed local field potential, appears to mainly reflect subthreshold activity, i.e., the synaptic input currents and their return currents. In general, recorded extracellular potentials are due to complicated weighted sums of contributions from transmembrane currents in the vicinity of the recording electrode. The interpretation of the recorded signals in terms of the underlying neural activity is thus often difficult. A biophysical forward-modeling scheme is available to aid the interpretation, however: transmembrane currents can be found from multicompartmental neuronal modeling, and may in turn be used to calculate extracellular potentials using an electrostatic forward-modeling formula based on the quasistatic version of Maxwell’s equations [1].

To facilitate such biophysical forward-modeling studies we have developed a Python toolbox aimed specifically for simulations of extracellular field potentials. It runs on top of the NEURON simulator (http://neuron.duke.edu/) and provides functions for simulation control and calculation as well as simple visualization of field potentials. With this software it is easy to set up simulations for studying, e.g., the dependence of recorded extracellular signature of action potentials on electrode position [2] or how the local field potentials generated by a synaptic input depends on neuronal morphology, synapse position and electrode position [3]. Using standard formats for neuronal morphologies, it allows for using morphological reconstructions available in online databases (e.g., http://neuromorpho.org/) while taking advantage of the wide range of powerful analysis and visualization tools written in the Python programming language .

[1] Holt & Koch, J Comp Neurosci 6:169 (1999)

[2] Pettersen & Einevoll , Biophys J 94:784 (2008)

[3] Lindén et al , J Comp Neurosci, 29:423 (2010)