Statistical analysis (coefficient of variation, PCA and ANOVA analysis)
Within a population the absolute variation or dispersion per trait is defined as the standard deviation (σ). The relative variation called the coefficient variation (CV) for individual traits is the ratio of the standard variation to the mean (µ) of the lines in the population (CV = (σ/µ)*100). In this study we calculated CV for each metabolite in two nutritional maternal environments separately. The metabolomics data were log10 transformed and then used for further analysis such as ANOVA, principal component and correlation analysis. In order to indicate the effect of the genotype, environment and their interaction, ANOVA analysis was performed on the metabolite content of the parental lines, MM and PI, grown in different nutritional environments. The significance threshold of the p -value was set to 0.05. Principal component analysis (PCA) was performed on metabolic values of the RILs and the parental lines using the R-package “pcaplots”.
Log2 ratio of metabolites between HP and LN (HP:LN) and metabolite profiles in both maternal environments (HP and LN) were investigated to identify the metabolic differences between two nutritional maternal environments with Metaboanalyst 3.0 (http://www.metaboanalyst.ca/faces/home.xhtml).