Computational protocol: Two-Photon Functional Imaging of the Auditory Cortex in Behaving Mice: From Neural Networks to Single Spines

Similar protocols

Protocol publication

[…] All the data were analyzed offline by custom-written software in LabVIEW 2014, Igor Pro 5.0 (Wavemetrics Inc., USA) and MATLAB 2016b (MathWorks, USA).To evaluate the performance of the mice in the sound-triggered licking task during training, the tongue movements were measured based on the video with custom-written LabVIEW software and the success rate of the licking response was quantified accordingly. The detection window for successfully licking during data analysis is set as 500 ms. In detail, a region of interest (ROI) was manually drawn in the video frame over the mouth and the water spout. The licking strength was calculated as the frame-by-frame difference of the image intensity of the ROI covered the mouth region of the mouse, hence the “arbitrary units” is just a relative unit of measurement to represent the amount of intensity change.As previous studies described (Li et al., ), the individual neurons in two-photon imaging data were identified visually, the ROIs for each neuron were drawn manually and the Ca2+ signal of each ROI was calculated as the relative fluorescence change Δf/f = (f-f0)/f0 over time. The f0 was estimated as the 25th percentile of the fluorescence values for each ROI. Detection of Ca2+ transient was performed based on thresholding criteria about peak amplitude and rising rate. We defined a noise level as 3 times the standard deviation (SD) of the baseline activity.For correcting brain motions resulting from movements during mouse behavior, we used the image alignment software TurboReg (ImageJ, NIH, USA). The brain motions of mice can be decomposed into motions that parallel to the focal plane (XY-plane) and perpendicular to the focal plane (Z-axis). However, Z-axis motion cannot be corrected offline, and the improvement of offline correction was resulted in the XY axis motions. Frame-by-frame alignment of the imaging data was performed with a translation algorithm (Supplementary Video and Supplementary Figure ). The brain motions were quantified by mean frame to frame in plane (XY) Euclidean distances. For naive mice (4 mice, 2 focal planes for each mouse, 60 s imaging time for each focus), the motion was 2.1 ± 0.9 μm (mean ± SD). After the same group of mice was habituated to head fixation, the motion was 1.9 ± 1.2 μm, and the brain motions larger than 4.5 μm were significantly reduced (Supplementary Figure ). Furthermore, the brain motion of the conditioned mice was 1.1 ± 0.6 μm (4 mice, 66 behavioral trials in total).For the reconstruction of the neuron morphology, Z-stack fluorescent images of a single neuron were projected into an averaged image by using projection software (ImageJ). Based on the averaged projections, a schematic morphology of the single neuron, including soma and dendrites, was manually drawn in Adobe Illustrator CS6 (Adobe Systems, USA). […]

Pipeline specifications

Software tools ImageJ, Adobe Illustrator
Applications Miscellaneous, Microscopic phenotype analysis
Organisms Mus musculus