The relative contribution of edaphic, topography, DBH and spatial to methylation rate
Variation-partitioning is a method using tables of various variables. By adjusting the coefficient of determination, partitioning provides an unbiased estimate in both multiple regression and canonical redundancy analysis on the variation of methylation level data table that is explained by the various tables of explanatory variables. In here, the multivariate variation-partitioning approach was conducted by using the R package “varpart” to quantify the relative contributions of microhabitat, individual life stage and spatial attributes, as well as their potential intersections, to the variation of methylation rate. Before the variation partitioning analysis, the variables most significantly influential to methylation level were chosen by permutational forward model selection by using the method developed previously (Chen et al ., 2016) with the R package “packfor”.