Fig. 3: Mesmerize handles 3D calcium imaging data.

a Mesmerize can work with volumetric calcium imaging data. b Frame from one plane of the volumetric dataset. CNMF with greedy initialization is unable to detect many cells in this relatively noisy dataset; however, CNMF seeded with NuSeT segmentation picks up many more cells. c Mesmerize datapoint tracer showing a cell highlighted in red, and the corresponding calcium trace. The tuning curves of this cell are shown in (d), which shows that this cell is tuned to heat-on stimulus. e Min−max normalized calcium traces sorted by their stimulus tuning profiles, heat-on, heat-on-delayed, and none. Color bar indicates normalized fluorescence intensity. f Spacemap showing the stimulus tuning characteristic of each cell. g LDA projection showing distinct brain states for heat-on, heat-on-delayed and none between each stimulus trial. Scale bars: 100 μm. heat-on: cells that respond to the heat stimulus; heat-on-delayed: cells that show a delayed response to the heat stimulus; none: cells that are more active between the stimulus trials and less active during heat-on and heat-on-delayed stimulus periods.