Sas version 9

Distribution of the platelet, leukocyte, growth factor, and cytokine concentrations were assessed for normality using histograms, probability plots, and multiple tests for normality including a Shapiro–Wilk test (SAS version 9.4, Cary, NC, USA). The platelet and total leukocyte data were adequately normal to justify comparison among PRP systems using a linear mixed effects model (SAS version 9.4, Cary, NC, USA). The full model included a fixed factor for the system and a random intercept for each dog. Pair-wise comparisons were made using a Tukey’s test.
Data for TGF-β1 and PDGF-BB concentrations were not normally distributed so differences among PRP preparations using different systems were compared using a non-parametric Friedman test (SAS version 9.4, Cary, NC, USA). When significant, multiple comparisons were performed using Conover’s test [R (R Core Team); http://www.R-project.org/]. A Bonferroni correction was used to adjust for multiple comparisons.
Correlation between platelet concentration and TGF-β1 concentration was assessed for those PRPs that were not activated (Systems 1, 3, 4, 5). Correlations between platelet concentration and both TGF-β1 and PDGF-BB concentrations were assessed for the PRP releasates from System 2 that had been activated. All correlations were assessed using a Pearson’s test.

In Exp. 1 and Exp. 2, the phenotypic data measured on leaf physiological traits (Chlorophyll content, stomatal content, osmotic potential, and leaf water potential), shoot and root fresh, and dry biomass were analyzed using the GLM approach with “blocks” as a fixed variable. Step-wise regression and correlation analyses between traits measured were also performed using SAS version 9.1. The image-based analysis of alfalfa three leaflet blade (area, length, width, entropy, and compactness) and root (diameter, volume, width, angle, and vertical root length) traits were performed using methods described earlier (Liao et al., 2017 (link); Paez-Garcia et al., 2019 (link)). Analysis of variance (ANOVA) of shoot (NDVI) and root traits (median, maximum number of roots, total root length, max width, network area, solidity, perimeter, average radius, volume, surface area, maximum radius, steep angle frequency, and holes) from raised-bed experiment was analyzed using ProcGLM analysis in SAS version 9.1, and their relationships between traits were performed using SAS JMP version 15 program. In all the experiments, the least significant difference (LSD) at α = 0.05 was used to determine significant differences among genotypes, treatments, and the interaction. Statistical significance was based on a p-value of 0.05.

Data on the microorganism population, chemical composition and fermentation quality after 60 d of ensiling were analyzed with a completely randomized design with a 2×4 (sealing [S] ×additive [A]) factorial treatment structure. The two ways analysis of variance (ANOVA) procedure of SAS version 9.1 (SAS Institute, Cary, NC, USA) was used for the analysis and the statistical model is as follows:
where Y_ijk = observation; μ = overall mean, α_i = S treatment effect (i = Napier grass), β_j = A effect (j = 1 to 4), αβ_ij = S×A effect, and ɛ_ijk = error. The mean values were compared by Tukey’s test [18 ].
Data of aerobic stability on the lactic acid content, pH, temperature and counts of aerobic acid, yeast and mold after 1 d, 3 d, and 6 d of aerobic exposure were analyzed with a completely randomized design with a 2×3 (S×exposure days [D]) factorial treatment structure. The two ways ANOVA procedure of SAS version 9.1 (SAS Institute, USA) was used for the analysis.

The Shannon index, ACE and Chao 1 data and fungal taxa abundance were analysed using GraphPad Prism 5.0 (GraphPad Software Inc., San Diego, CA, USA), and the significant differences in the microbial alpha diversity between different cadmium treatments were calculated using Duncan's multiple test at p<0.05, p<0.01 or p<0.001, respectively. Total and available cadmium were calculated using SAS version 9.1(SAS Institute Inc., Cary, NC, USA). One-way-analysis of variance (ANOVA) and Spearman's correlations were also calculated using SAS version 9.1(SAS Institute Inc., USA). To determine the potential differences in the bacterial and fungal communities across compartments and chromium-added treatments, a heatmap of Spearman's rank correlation was generated using the package “gpplot” (Oksanen et al., 2012) in R (v3.6.2) based on Bray-Curtis dissimilarity. Canonical discriminant analysis (CDA) was performed to calculate the environmental factors that have the most remarkable influence on the microbial and fungal communities by CANOCO 5.0 (Biometrics Wageningen, Netherlands) [72 , 73 ]. LEfSe analysis was applied using the version at https://huttenhower.sph.harvard.edu/galaxy/ [74 (link)].

The phenotypic data were analyzed using the general linear model for randomized complete block design using SAS version 9.1 [10 ]. Genome-wide association analyses were performed according to the method described by Aulchenko et al [11 (link)]. At first, slaughter-year-season were fit as a fixed effect and age of month as a covariate, using the SAS general linear model procedure in SAS version 9.1. After that, the fixed or covariate having 0.1 statistical significance level was fit into a mixed model with a genome-relationship matrix (G matrix) [12 (link)], as the pedigree information of the 140 steers was limited. The G matrix was constructed using an R subroutine (version 2.15.0), and the residuals of each phenotype were obtained from the equation of mixed model using ASREML (version 3.0). Secondly, the residuals were regressed on each SNP using a simple linear regression model using PLINK version 1.07. For each individual, SNP genotypes for BB, BA, and AA were defined as −1, 0, and 1, respectively, such that allele substitution effect by replacing B with allele A was estimated for each SNP. To set threshold values of statistical significance, 0.1% point-wise p-value from the F distributions was applied for each SNP test.

Prior to statistical analysis, percentage data were arcsine transformed, and count data were square-root transformed to ensure the normality of the data. The differences in H. armigera oviposition preference on a multiple-choice experiment, no- choice experiment, volatile experiment, and oviposition on tomato and S. viarum mixture were analyzed by analysis of variance (ANOVA) with the Proc GLM procedure of SAS version 9.1 (SAS Institute, Cary, NC, USA). The two-choice oviposition experiment of H. armigera oviposition on S. viarum accessions and tomato were analyzed by paired t-test using online software GraphPad Prism 8. Differences in evaluated plant morphological (trichomes) and chemical (acylsugars and phenolics) characteristics and life parameters of the insect were analyzed by analysis of variance (ANOVA) with the Proc ANOVA procedure of SAS version 9.1 (SAS Institute, Cary, NC, USA). Mean separation was completed using Tukey’s Honest Significant Differences (HSD) post-hoc test and significant treatment differences were indicated. Finally, the correlations between different plant and insect life parameters were analyzed by IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp, Armonk, NY, USA).