SUMMARY In the vertebrate visual system, all output of the retina is carried by retinal ganglion cells. Each type encodes distinct visual features in parallel for transmission to the brain. How many such “output channels” exist and what each encodes is an area of intense debate. In mouse, anatomical estimates range between 15–20 channels, and only a handful are functionally understood. Combining two-photon calcium imaging to obtain dense retinal recordings and unsupervised clustering of the resulting sample of >11,000 cells, we here show that the mouse retina harbours substantially more than 30 functional output channels. These include all known and several new ganglion cell types, as verified by genetic and anatomical criteria. Therefore, information channels from the mouse’s eye to the mouse’s brain are considerably more diverse than shown thus far by anatomical studies, suggesting an encoding strategy resembling that used in state-of-the-art artificial vision systems.

Summary A fundamental challenge in calcium imaging has been to infer spike rates of neurons from the measured noisy fluorescence traces. We systematically evaluate different spike inference algorithms on a large benchmark dataset (>100.000 spikes) recorded from varying neural tissue (V1 and retina) using different calcium indicators (OGB-1 and GCaMP6). In addition, we introduce a new algorithm based on supervised learning in flexible probabilistic models and find that it performs better than other published techniques. Importantly, it outperforms other algorithms even when applied to entirely new datasets for which no simultaneously recorded data is available. Future data acquired in new experimental conditions can be used to further improve the spike prediction accuracy and generalization performance of the model. Finally, we show that comparing algorithms on artificial data is not informative about performance on real data, suggesting that benchmarking different methods with real-world datasets may greatly facilitate future algorithmic developments in neuroscience.

SummaryPrimary visual cortex exhibits two types of gamma rhythm: broadband activity in the 30–90 Hz range and a narrowband oscillation seen in mice at frequencies close to 60 Hz. We investigated the sources of the narrowband gamma oscillation, the factors modulating its strength, and its relationship to broadband gamma activity. Narrowband and broadband gamma power were uncorrelated. Increasing visual contrast had opposite effects on the two rhythms: it increased broadband activity, but suppressed the narrowband oscillation. The narrowband oscillation was strongest in layer 4 and was mediated primarily by excitatory currents entrained by the synchronous, rhythmic firing of neurons in the lateral geniculate nucleus (LGN). The power and peak frequency of the narrowband gamma oscillation increased with light intensity. Silencing the cortex optogenetically did not abolish the narrowband oscillation in either LGN firing or cortical excitatory currents, suggesting that this oscillation reflects unidirectional flow of signals from thalamus to cortex.

In recent years, two-photon calcium imaging has become a standard tool to probe the function of neural circuits and to study computations in neuronal populations. However, the acquired signal is only an indirect measurement of neural activity due to the comparatively slow dynamics of fluorescent calcium indicators. Different algorithms for estimating spike rates from noisy calcium measurements have been proposed in the past, but it is an open question how far performance can be improved. Here, we report the results of the spikefinder challenge, launched to catalyze the development of new spike rate inference algorithms through crowd-sourcing. We present ten of the submitted algorithms which show improved performance compared to previously evaluated methods. Interestingly, the top-performing algorithms are based on a wide range of principles from deep neural networks to generative models, yet provide highly correlated estimates of the neural activity. The competition shows that benchmark challenges can drive algorithmic developments in neuroscience.