Matlab r2015

Univariate statistical analysis (UVA) was carried out with an in-house developed script for MATLAB R2015 (Mathworks, Inc., Natick, MA, USA) whereas SIMCA P + 12.0.1 was used for multivariate analyses (MVA). For MVA, log-transformation and Unit Variance (UV)-scaling were used to build the respective matrices per technique. Unsupervised (Principal Components Analysis, PCA) and supervised (orthogonal partial least squares discriminant analysis, OPLS-DA) analyses were applied to check trends, outliers, and to select discriminating variables correspondingly. For UVA, Repeated Measures ANOVA was applied. The factors were the development of atherosclerosis (“within factor”, associated with time) and the treatment with colchicine (“between factor”, associated with the group). Bonferroni-corrected p-values for the significance of each factor, as well as for the interaction, were computed for all metabolites. Fold change (FC) was calculated as (average value in the case group/average value in the reference group). For those metabolites that showed p-value < 0.05 for any factor or interaction, two log₂(FC) were calculated: of the comparison of [rabbits after 36 weeks vs. rabbits after 18 weeks, both without colchicine] (i.e., atherosclerosis progression) and [rabbits after 36 weeks with colchicine vs. rabbits after 36 weeks without colchicine] (i.e., colchicine treatment).

The baseline correction and phasing of NMR spectra were performed with TopSpin software (Bruker, USA) for chemical shift referencing. NMR spectral data were processed using MATLAB R2015 version 9.10 (MathWorks Inc., USA) software equipped with the IMPaCTS toolbox (https://doi.org/10.5281/zenodo.3077413) for conducting probabilistic quotient normalization (PQN). Statistical total correlation spectroscopy (STOCY) [11 (link)] was used to validate the appearances of correlated resonances on 1-dimensional NMR spectra which were searched for against public databases including the human metabolome database (HMDB) and ChenomxNMR Suite version 9.0 (Chenomx Inc., Canada) for metabolite identification. Data can be accessed at Open Science Framework (https://osf.io/sqftm/).

Deviations between the digitized master model and the stone model scans were analyzed with the aid of Geomagic Design X (3D Systems, Germany) and Matlab R2015 (Mathworks, Natick, MA, USA). Scan regions showing the precision balls were cropped, sphere center positions calculated by means of optimization (least squares, fixed nominal sphere radius), and sphere distances determined. After use of a coordinate transformation aligning the global scan coordinate system with that of the reference model, each of the three reference surfaces was separately aligned with each scan and the local coordinate systems moved along (cf. Fig. 2). In detail, the following measurement data (n = 10 samples per subgroup, 3 measurement repetitions per sample) were reported in this manuscript:

Global accuracy

Distance changes between precision ball centers

Distance changes between preparation margin centers and angular changes between the tooth axes

Local accuracy

For each prepared surface, trueness and precision were analyzed separately (based on absolute values) by means of mesh deviation.

The data were digitally filtered using a low-pass filter with a cut-off frequency of 0.67 Hz to eliminate cardiac-related impedance changes. The data were analyzed offline using customized software programmed with MATLAB R2015 (the MathWorks Inc., Natick, MA, USA). EIT images were divided into four symmetrical, non-overlapping and ventral-to-dorsal horizontal regions of interest (ROIs), ranging from the gravity-independent areas to the gravity-dependent areas, namely, the ventral (ROI1), mid-ventral (ROI2), mid-dorsal (ROI3), and dorsal (ROI4) regions. Moreover, the ventilation maps were also divided into symmetrical, non-overlapping, four cross quadrants: lower left (LL), lower right (LR), upper left (UL) and upper right (UR). TV/VT was defined as the ratio of tidal impedance variation/tidal volume.

The following criteria were used for inclusion of a variant in our analyses: a minimum proportion of the variant within the sample of 1% (to mitigate the effects of amplification and sequencing errors [13 (link),14 (link),35 (link),36 (link)], and a minimum of 5 sequence reads for the variant. Sensitivity analyses using different threshold settings are shown below under “Sensitivity analyses”. This data selection and all subsequent analyses were performed in Matlab R 2015 (The Mathworks) unless otherwise indicated.

All patients were studied using either a 1.5- or 3.0-T MRI scanner prior to surgery. T1-weighted imaging with and without gadolinium enhancement, T2-weighted (T2WI), and Fluid attenuation inversion recovery (FLAIR) images were acquired in all cases for delineation of tumors. T2WI was used for further analysis. Digital Imaging and COmmunication in Medicine (Dicom) format images were converted into Neuroimaging Informatics Technology Initiative (NIfTI) format using MRIConvert (http://lcni.uoregon.edu/downloads/mriconvert) and were subsequently analyzed by an in-house software developed on Matlab R2015 (MathWorks, Natick, MA). The developed software is capable of creating voxels-of-interest (VOI) in three dimensions from the original NIfTI data by manual segmentation. High intensity lesions on T2WI were manually segmented to create a 3D VOI for all of the cases (Fig 1). Diffuse high intensity lesions surrounding the presumed tumor core were all included in the VOI.