Simca p version 13

Visualization of the data and data analysis were carried out with MeV 4.9.0 software (Saeed et al., 2003 (link)). Every experiment was carried out in triplicate. Statistical significance from different treatments was revealed after one-way analysis of variance (ANOVA) followed by Tukey’s test. Partial Least Square (PLS) models were performed using SIMCAP+ version 13.0 (Umetrics AB, Umeå, Sweden) with 14 metabolites as X variables and eight biological activities as Y variables. All variables were mean-centered and unit-variance (UV) scaled prior to PLS. Correlation analysis was performed using Past 3.0 (Øyvind Hammer, Natural History Museum, University of Oslo, Oslo, Norway) using the Pearson parametric correlation test and visualized using Heatmapper (Babicki et al., 2016 (link)). Significant thresholds at p < 0.05 with significant differences represented by different letters or p < 0.05, <0.01, and <0.001 were used for all statistical tests and represented by different letters or by *, **, and ***, respectively.

For analysis of demographic data, the median value (min-max) was calculated. Wilcoxon paired test was used to compare results before and after the intervention. These statistics were performed using the statistical package IBM SPSS Statistics (version 24.0; IBM Corporation, Route 100 Somers, New York, NY, USA). A probability of <0.05 (two-tailed) was accepted as the criteria for significance. To investigate the multivariate correlation and between membership of groups and quantified proteins and metabolites orthogonal partial least square discriminant analysis (OPLS-DA) was used (SIMCA-P+ version 13.0, UMETRICS, Umeå, Sweden). Before this analysis, principal component analysis (PCA) was used to check for multivariate outliers. The procedure to compute multivariate correlation models has been described earlier [41 (link)] and is in accordance with [42 (link)].

The resulting data sets were then imported into SIMCA-P version 13.0 (Umetrics, Umeå, Sweden) for further multivariate data analyses.
Prior to PCA, the bucket data were mean-centered and then scaled using Pareto scaling. Hotelling’s T² region, shown as an ellipse in the score plots, defined the 99% confidence interval of the modeled variation. The quality of the model was described by Rx² and Q² values. Rx² was defined as the proportion of variance in the data explained by the model, indicating goodness of fit. Q² was defined as the proportion of variance in the data that was predictable by the model, indicating predictability.
OPLS-DA provided the maximum covariance between the measured data (X) and the response variable (Y). The response variables represent the intensities of the corresponding buckets assigned to separate components. The overall predictive ability of the model was assessed by cumulative Q², which represented the fraction of the variation in Y that can be predicted by the model⁴⁷ .

Temporal changes in BPI3V antibody titre were analysed using two-tailed paired T-test, and significant differences between treatment groups at sampling stages was assessed using two-tailed heteroscedastic T-test. SIMCA-P+ version 13.0 (Umetrics, Sweden) was used for multivariate metabolite marker selection. For SIMCA-P+ analysis, data was prefiltered to exclude AMRTPs with coefficient of variation greater than 50% in replicate quality control pools. All centroid data were Pareto scaled and analysed by unsupervised Principle Component Analysis (PCA) and supervised discriminatory analysis by Orthogonal Projections of Latent Structures-Discriminant Analysis (OPLS-DA). Unsupervised PCA models were generated at each sampling day to reveal potential relationships between treatment groups. Supervised analysis by OPLS-DA was performed to reveal potential markers of response to treatment in vaccinated calves compared to non-vaccinated calves at each sampling day. Robustness of final OPLS-DA discriminative models was assessed by setting a predictive model of each case in which 2/3 of the data (known treatment) was used to predict the remaining 1/3 (unknown treatment).

The relative data acquired from polar metabolites profiling were subjected to orthogonal projections to latent structures discriminant analysis (OPLS-DA; SIMCA-P version 13.0; Umetrics, Umeå, Sweden) to inspect which components separated between green and purple pakchoi samples [33 (link)]. The OPLS-DA output consisted of score plots for visualizing the contrast between different samples and loading plots to explain the cluster separation. The data file was scaled with unit variance scaling before all the variables were subjected to OPLS-DA. Statistical analysis was performed with SPSS statistics 22 software (IBM Corp., Armonk, NY, USA) using Student’s t test. All data are given as the mean values and standard deviation of triplicate experiments.