To select a dataset for display, click on a brain area and then click on a task behavior. Select one or more ephys or imaging recordings. Click “Display”. You can also add one or more spike-to-imaging or imaging-to-spike models (the models will be automatically matched to the selected datasets).

Ephys

Imaging



Spike to imaging models

Imaging to spike models




Benchmark table
Dataset Name Fraction selective neuron Peakness PC content Decodability


Distribution table
Dataset Name Fraction selective neuron Peakness PC content Decodability



By simultaneously recorded data, we refer to a recording that provides both a direct measurement of the spiking activity of a neuron and that of the indirect method of interest. In our case simultaneous spiking and Ca imaging, as can be achieved for instance by combining loose-cell recordings and imaging.
Such datasets are far too rare in neuroscience. This resource collects many of them (Datasets) and we will gladly incorporate any submitted simultaneous dataset.
The first thing one can do with such a dataset is to develop a spike-to-calcium model that tests our understanding of the way calcium imaging is related to spiking activity by generating a predicted synthetic calcium trace given a set of spike times.
Details of each model are described in the Models section of this resource. Below we display quantifications and visualizations of the performance of models on the different simultaneously recorded datasets.
* Note: all models listed in Models are tested, and a small fraction of them with very low performances are not presented here. Nevetheless one can assess them using the full released API.

Model fits





Performances

All datasets

.0
.1
.2
.3
.4
.5
.6
.7
.8
.9

Single datasets (click label in table to see dataset name)

.0
.1
.2
.3
.4
.5
.6
.7
.8
.9
Model Explained variance Spearman's correlation
S2C Linear
S2C-L
.60
.60
.71
.56
.51
.77
.76
.85
.77
.71
S2C Hill
S2C-H
.68
.68
.77
.61
.62
.82
.81
.89
.80
.78
S2C Sigmoid
S2C-S
.67
.67
.75
.60
.61
.81
.81
.88
.80
.78
C2S Nonnegative Wienner Filter
NWF
.30
.33
.73
-.04
.05
.86
.83
.96
.90
.80
C2S FOOPSI
FOOPSI
.77
.70
.95
.90
.63
.87
.83
.97
.92
.79
C2S Finite Rate Innovation
FRI
-1.4
-1.6
-.95
-2.3
-.88
.60
.56
.68
.60
.59
C2S Constrained OOPSI AR1
AR1
.72
.73
.66
.71
.74
.85
.86
.83
.82
.87
C2S Constrained OOPSI AR2
AR2
.77
.76
.78
.81
.76
.88
.87
.89
.87
.88
C2S Constrained OOPSI AR3
AR3
.76
.75
.80
.78
.72
.87
.87
.91
.85
.85
C2S Constrained OOPSI MCMC
MCMC
.80
.77
.86
.84
.76
.88
.87
.93
.88
.86
C2S Peel Linear
Peel
.50
.26
.60
.79
.61
.80
.68
.85
.86
.78
C2S MLSpike
MLS
.48
.22
.74
.79
.45
.63
.40
.87
.90
.63