Matlab software

The background signal from the samples containing no rhBMP-2 was subtracted from the absorbance value of each test sample. The bioactivity of the sample was determined by fitting the data using a four-parameter logistic equation model, where
Y = measured optical density at 405 nm,
X = concentration of rhBMP-2,
A = minimum optical density,
B = slope at the inflection point,
C = ED50 (concentration that reflects that at half of the maximum concentration), and
D = maximum optical density.
MATLAB software (version R2016b; Mathworks) was used for the curve fitting using a modified program.^* The upper limits of parameters A, B, C, and D were set at 1, infinity, 1369, and infinity, respectively. Lower limits were not set.

T₂ relaxation measurements were performed using a low-field nuclear magnetic resonance (LF-NMR) analyzer minispec PQ 001 (Niumag, Ltd., Shanghai, China). The fish sample (2 × 2 × 2 cm) was put into the detecting tube (70-mm diameter). Using the Carr-Purcell-Meiboom-Gill sequence to set the T₂ measurement parameters (9 (link)), the analysis software (NMI20-030H-1 NMR analyzer: Suzhou Niumag Analytical Instrument Co., Suzhou, China) was used to iteratively invert the collected signals to obtain the transverse relaxation time T₂ spectrum. Similarly, the proton density distribution of the sample is determined by magnetic resonance imaging (MRI) after setting the parameters. The gray-scale map of proton intensity was altered to be pseudocolor images by MATLAB software (MathWorks Inc., Natick, MA, USA).

MRI process and analysis were performed as previously described.
²³ (link) For VBM analysis, all mice 3D T2‐WI images were preprocessed using SPM12 software (Wellcome Department of Cognitive Neurobiology, University College of London, UK), which included linear registration, segmentation, normalization, Jacobian modulation, and smoothing.
For rs‐fMRI analysis, we first discarded 10 volumes of fMRIs. After correcting the slice timing and head motif for the rest of the volumes, they were spatially normalized to the template established by 3D T2‐WI images. Then, to smooth the images, an 8‐mm full width at half maximum (FWHM) Gaussian kernel was used. Signal regression and motion vectors were used to correct for systemic noise. Next, the seed region of interest (ROI) was obtained from the VBM results. Second, the resting‐state functional connectivity map between the PFC and the reward system was calculated using the Pearson correlation coefficient. Fisher Z‐transformation was performed to increase the normality of the functional connectivity data. After statistical analysis of functional connectivity values using GraphPad Prism 8 (La Jolla, CA, USA), t‐values reflecting the differences in functional connectivity were converted into a matrix using MATLAB software (The MathWorks Inc).

The HSIs in the SWIR and Vis-NIR ranges were segmented using the Otsu technique [38 (link)]. During the segmentation process, we developed binary images, with the pixels of the unwanted background set at zero (black), and the nonzero (white) elements were utilized to extract the pixels of the bread samples. The average of the spectra of the pixels of the bread samples for each image was obtained to acquire a representative spectrum. For the Vis-NIR region, images at 526 nm were utilized to segregate the pixels of concern from the background. For the SWIR region, the best contrast between the pixels of the bread sample and their background was obtained at 1513 nm. Also, the wavebands with the smallest amount of signal-to-noise ratio were removed. The binary images were applied to all the 224 and 288 waveband images for the Vis-NIR and SWIR wavelength regions, respectively. The spectral data obtained from each bread sample was arranged in an Excel file for multivariate data analysis. MATLAB software (MATLAB version 9.6, R2019a, MathWorks Inc., Natick, MA, USA) was used for the image segmentation and spectral data extraction [32 (link)].

The fluorescence imaging of the Ca²⁺ dynamics of NSCs cultured on Au nanostrip array with a rotating magnetic field (300 rpm) for 5 days was performed using a fluorescent Ca²⁺ indicator (Fluo‐4 AM, Beyotime Biotechnology). Briefly, neurons on different substrates were washed with PBS three times and incubated with 0.5 × 10⁻⁶ m Fluo‐4 AM for 10 min at 37 °C in HEPES (Sigma‐Aldrich, USA). After incubation with Fluo‐4 AM, the neurons were washed with PBS and incubated for another 20 min to allow complete de‐esterification of the dye. A LSM800 confocal laser scanning microscope (Carl Zeiss) was used to observe and photograph the neurons labeled with Fluo‐4 AM. Neurotransmitters (acetylcholine, g‐aminobutyric acid, dopamine, or glutamine) at a concentration of 500 × 10⁻⁶ m were used to stimulate neurons to trigger Ca²⁺ sparks. The relative fluorescence intensity change (%ΔF/F) for individual neurons showing the change in intracellular concentration of Ca²⁺ after stimulation by the neurotransmitters was calculated using MATLAB software (MathWorks, USA). The fluorescence imaging of the Ca²⁺ dynamics of the NSCs cultured on TCP without a rotating magnetic field (300 rpm) for 7, 9, or 10 days was performed in the same way.