Mandatory Publication of Computational Brain Models Simultaneously with Paper!

In yesterday’s blog post “Brain Modeling Using NEURON, Interneurons, and Resonant Circuits” I reviewed some of the results of a new computational neuroscience paper “Computational modeling of distinct neocortical oscillations driven by cell-type selective optogenetic drive: separable resonant circuits controlled by low-threshold spiking and fast-spiking interneurons” (published November 22, 2010 in Frontiers in Human Neuroscience). The research team used computational models of the cerebral cortex that ran in the NEURON simulation environment. In the paper’s methods it’s stated that “Upon publication, the model will be made available in the model DB database.” My first thought was “This is great! This is how research in computational neuroscience should be published.” My assumption was that the model or models presented in the paper would be available at ModelDB. Unfortunately I was disappointed.

Note: ModelDB is the computation neuroscience model repository within the larger SenseLab online data repository.

The paper references two NEURON models in the ModelDB online data repository. One is “Pyramidal Neuron Deep, Superficial; Aspiny, Stellate (Mainen and Sejnowski 1996)” described in the paper “Influence of dendritic structure on firing pattern in model neocortical neurons” published July 25, 1996 in Nature and the other is “MEG of Somatosensory Neocortex (Jones et al. 2007)” described in the paper “Neural Correlates of Tactile Detection: A Combined Magnetoencephalography and Biophysically Based Computational Modeling Study” published October 3, 2007 in the Journal of Neuroscience. Neither of these is one of the actual models that are used in the paper under review.

The cited models include the “canonical” model neurons and network components that are used in the current paper. The authors seem to include all of the parameters, etcetera that are needed to extend and modify the canonical models into the NEURON models under discussion. While this is a step forward from no model repository it falls short of the ideal where experimental neuroscientists, students, and others could quickly play with a theorist’s model in conjunction with reading the paper.

Funding institutions where are you? Make it mandatory to publish the model or models described in a paper simultaneously when publishing the paper itself!

Other related blog posts:

Brain Modeling Using NEURON, Interneurons, and Resonant Circuits