Computational Models of Astrocytes and Astrocyte–Neuron Interactions: Characterization, Reproducibility, and Future Perspectives

Abstract

Astrocytes have been shown to participate in a variety of brain functions. These include homeostasis, metabolism, neuronal survival in pathological circumstances, and neurovascular coupling. Since astrocytes extend their processes into close proximity to synapses, it has also been proposed that they take active roles in synaptic transmission, learning, and memory. The complexity of dynamic interactions on both molecular and cellular levels of neurons and astrocytes is overwhelming. This underlines the demand for detailed, integrative computational models for advancing our understanding of the functional contribution of astrocytes in the nervous system. This study presents the state of the art in computational models for astrocytes and astrocyte–neuron interactions. First, we characterized the models based on the type of biological entities they described. We then studied several aspects of the models in detail, including reproducibility. We discovered that several publications lack crucial details in how the models were presented, preventing successful reproduction of the results. Graphical illustrations of these models were misleading, mathematical equations incorrect, or selected model components not adequately justified. Moreover, in some cases, it was impossible, after several trials, to reproduce the simulated results presented in the original publications. In order to facilitate reproducible science, we propose some criteria that computational glioscience models should meet. To the best of our knowledge, this study is one of the first to report the detailed categorization and evaluation of astrocyte-neuron models.