Skip to main content
Fig. 2 | BMC Biology

Fig. 2

From: Neuron ID dataset facilitates neuronal annotation for whole-brain activity imaging of C. elegans

Fig. 2

Variations of cell positions. a Visualization of the variation of cell positions. The ellipsoid indicates the mean and the covariance of the positions of the cells. Cells in the right half of the body are shown. The colors are assigned randomly for visualization. In the case of the cells whose covariance cannot be calculated, the median of other covariance was used for visualization and shown in gray color. A/P means anterior-posterior, D/V means dorsal-ventral, and L/R means left-right directions. b Minimum distance (Euclid distance of centers of nearest ellipsoids) and the shortest axis length of the ellipsoids (equal to the twice of the smallest standard deviation) for each cell. The line shows where the minimum distance equals the shortest axis length. c Variation of the relative position of cell pairs is shown as a heat map. The red box and red dotted box indicate clusters of less varying cell pairs in the lateral ganglion and pharynx, respectively. For visualization, the variations were divided by their median value, and the color axis was truncated at 5 (the colors for cell pairs whose variation is larger than 5 are the same as the color for cell pairs whose variation is 5). d The error rate of the naive estimation method is visualized with cell positions in 3D. In the naive estimation method, the posterior probability of assignments was calculated for the respective cells in the respective animals based on the mixture of the Gaussian distributions. The name of the cell was estimated as the name of the Gaussian which had the largest probability for the cell. The error rates were calculated for each ground-truth cell. The hot color indicates that the error rate is high

Back to article page