Search

Recent development of large-scale activity imaging of neuronal ensembles provides opportunities for understanding how activity patterns are generated in the brain and how information is transmitted between neurons or neuronal ensembles. However, methodologies for extracting the component properties that generate overall dynamics are still limited. In this study, the results of time-lapse 3D imaging (4D imaging) of head neurons of the nematodeC. eleganswere analyzed by hitherto unemployed methodologies.By combining time-delay embedding with independent component analysis, the whole-brain activities were decomposed to a small number of component dynamics. Results from multiple samples, where different subsets of neurons were observed, were further combined by matrix factorization, revealing common dynamics from neuronal activities that are apparently divergent across sampled animals. By this analysis, we could identify components that show common relationships across different samples and those that show relationships distinct between individual samples.We also constructed a network model building on time-lagged prediction models of synaptic communications. This was achieved by dimension reduction of 4D imaging data using the general framework gKDR (gradient kernel dimension reduction). The model is able to decompose basal dynamics of the network. We further extended the model by incorporating probabilistic distribution, resulting in models that we call gKDR-GMM and gKDR-GP. The models capture the overall relationships of neural activities and reproduce the stochastic but coordinated dynamics in the neural network simulation. By virtual manipulation of individual neurons and synaptic contacts in this model, information flow could be estimated from whole-brain imaging results.

Additional file 10: Figure S7. Position of posterior pharyngeal bulb affects cell positions. (A-C) A/P (A), D/V (B) and L/R (C) positions of dye positive cells plotted against relative A/P positions of pharynx. The relative A/P positions of pharynx were calculated from the mean difference of positions of pharyngeal cells from reference. Blue crosses indicate the cell positions in respective animals. The red lines and the red dotted lines indicate regression lines and 95% confidence bounds, respectively. The summary statistics are in Additional file 11: Table S4.

Additional file 25: Figure S17. Most valuable cells for improving accuracy of automatic annotation. (A) Improvement of correct rate for prior annotated cell itself. Difference of correct rate is 0.9 when the correct rate of the cell in original case (i.e. without prior annotation) is 0.1. (B) Mean improvement of correct rate for the cells other than the prior annotated cells. Prior annotation of SMDVL achieved about 0.59% improvement of correct rate, indicating that 1.1 cells were corrected in addition to SMDVL. (C) Mean improvement of correct rate through all cells. Prior annotation of SMDVL achieved about 1.1% improvement of correct rate, indicating that 2.0 cells were corrected.

Additional file 5: Figure S3. Movements of the cells during time-lapse imaging. (A) The mean position and covariance of cell positions before the translation correction are shown as ellipsoids (see Fig. 2). An adult animal of JN3038 strain (see below) was introduced in the customized olfactory chip (see Methods) and imaged for about 20 min (6000 volumes). The nuclei in the volumetric movie were detected and tracked. Note that the origins of the axes are the same as those of the obtained raw images and the cell positions cannot be compared directly to other data including Fig. 2a. (B) The mean position and covariance of cell positions after the translation correction are shown as ellipsoids. The volumes of ellipsoids are smaller than that in Fig. 2a, indicating that the temporal movements alone cannot explain the large variations of cell positions shown in Fig. 2. The summary statistics are in Additional file 6: Table S3.

Additional file 8: Figure S5. Specific-cell-centered landscape. (A) Original landscape as a reference. This panel is basically the same as Fig. 2a, but several cells are removed for visibility. (B) ASKR-centered landscape. The position of ASKR cell is indicated as a cross. (C) MI-centered landscape. The position of MI cell is indicated as a cross. The same cell has the same color in (A)-(C). The cells in the right side are shown. Several cells are removed for visibility.

Additional file 7: Figure S4. Overlay plot of cell positions for all worms. (A) The positions of cells in the left half of the body for all worms are plotted. Colored circles indicate the positions of identified cells. Gray circles indicate the positions of unidentified cells. Different colors mean different identities. (B) Same as (A) but only for identified cells. (C) Same as (A) but only for identified non-pharyngeal cells. (D) Same as (A) but only for identified pharyngeal cells.

Additional file 9: Figure S6. Less varying neuron pairs. (A) Less varying neuron pairs were obtained by random permutation of animals (see Methods) and the less varying pairs in the left half of the body are shown by red lines. The pairs including pharyngeal cells were omitted for visualization. (B) The less varying pairs including pharyngeal cells are shown by red lines.

Additional file 3: Figure S2. Performance and robustness of the posture correction. (A) The variation of cell positions was evaluated as the volume of ellipsoid (corresponding to the determinant of covariance of cell positions; see Fig. 2) before and after the posture correction. The variation of cell positions was plotted against the mean cell position in the anterior-posterior (AP) axis (circles and crosses). Note that the y axis is in logarithmic scale. The lines indicate moving average obtained with a window of ±15 μm. (B) The moving average of variation of cell positions after posture correction steps that consist of principal component analysis (PCA), quadratic curve fitting, rotation, and translation (see Methods), each of which reduced the variations. (C) Up to 20% of the cells were randomly removed before the posture correction that simulates overlooking of cells in the nucleus detection step. The lines indicate the moving averages of variation of cell positions after the posture correction. The lines are overlapped, suggesting that the posture correction step is robust for the overlooking of cells. (D) Up to 20% of the cells either in the anterior side or in the posterior side were removed before the posture correction that simulates cells moving out of the view of the images. The side in which the cells were removed was randomly chosen in each animal. The lines indicate the moving averages of variation of cell positions after the posture correction. The lines are overlapped, suggesting that the posture correction step is robust for the movement of the cells. The summary statistics of (A)-(D) are in Additional file 4: Table S2. (E) The cell counts are shown against the mean cell position in the AP axis. The small counts of the cells in the posterior side might increase the instability of the moving averages of ellipsoid volumes in the posterior side. (F) The variation of cell positions is shown against the cell counts. The variation and the cell counts did not seem to correlate in general.

BackgroundAnnotation of cell identity is an essential process in neuroscience that allows comparison of cells, including that of neural activities across different animals. InCaenorhabditis elegans, although unique identities have been assigned to all neurons, the number of annotatable neurons in an intact animal has been limited due to the lack of quantitative information on the location and identity of neurons.ResultsHere, we present a dataset that facilitates the annotation of neuronal identities, and demonstrate its application in a comprehensive analysis of whole-brain imaging. We systematically identified neurons in the head region of 311 adult worms using 35 cell-specific promoters and created a dataset of the expression patterns and the positions of the neurons. We found large positional variations that illustrated the difficulty of the annotation task. We investigated multiple combinations of cell-specific promoters driving distinct fluorescence and generated optimal strains for the annotation of most head neurons in an animal. We also developed an automatic annotation method with human interaction functionality that facilitates annotations needed for whole-brain imaging.ConclusionOur neuron ID dataset and optimal fluorescent strains enable the annotation of most neurons in the head region of adultC. elegans, both in full-automated fashion and a semi-automated version that includes human interaction functionalities. Our method can potentially be applied to model species used in research other thanC. elegans, where the number of available cell-type-specific promoters and their variety will be an important consideration.

Annotation of cell identity is an essential process in neuroscience that allows for comparing neural activities across different animals. InC. elegans, although unique identities have been assigned to all neurons, the number of annotatable neurons in an intact animal is limited in practice and comprehensive methods for cell annotation are required. Here we propose an efficient annotation method that can be integrated with the whole-brain imaging technique. We systematically identified neurons in the head region of 311 adult worms using 35 cell-specific promoters and created a dataset of the expression patterns and the positions of the neurons. The large positional variations illustrated the difficulty of the annotation task. We investigated multiple combinations of cell-specific promoters to tackle this problem. We also developed an automatic annotation method with human interaction functionality that facilitates annotation for whole-brain imaging.