Estimation of functional connectivity of neuronal network from spike data

我估计了神经网络的功能连通性，从培养的神经元中记录的神经突数据。利用动态贝叶斯估计算法，得到了一个突触权值分布具有长尾的网络模型。所获得的模型能够再现神经元活动的爆发样，这种活动在突触权重被打乱后消失。这些结果表明，重量分布和网络结构都是诱发爆发样神经元活动的重要因素。

I estimated a functional connectivity of neuronal network from burst like neuronal spike data recorded from cultured neurons. With using a dynamic Bayesian estimation algorithm, I obtained a network model the distribution of synaptic weights of which has a long tail. The obtained model could reproduce the burst like neuronal activities, which disappeared when the synaptic weights were shuffled. These results suggest that both the weight distribution and the network structure are important to induce the burst like neuronal activities.