Principal Component Analysis

methrix_pca(
  m,
  var = "top",
  top_var = 1000,
  ranges = NULL,
  pheno = NULL,
  do_plot = TRUE,
  n_pc = 2
)

Arguments

m	Input `methrix` object
var	Choose between random CpG sites ('rand') or most variable CpGs ('top').
top_var	Number of variable CpGs to use. Default 1000 Set it to NULL to use all CpGs (which is not recommended due to memory requirements). This option is mutually exclusive with `ranges`.
ranges	genomic regions to be summarized. Could be a data.table with 3 columns (chr, start, end) or a `GenomicRanges` object
pheno	Column name of colData(m). Default NULL. Will be used as a factor to color different groups
do_plot	Should a plot be generated?
n_pc	Default 2.

Value

PCA results

Examples

data('methrix_data')
methrix_pca(methrix_data, do_plot = FALSE)
#> $PC_matrix
#>          PC1        PC2        PC3           PC4
#> C1 -1.539691  5.1212315 -0.3385356  1.079377e-15
#> C2 -6.255857 -2.7330127  0.2598231 -3.866346e-15
#> N1  3.774638 -1.5860181 -3.1984009  5.011880e-15
#> N2  4.020911 -0.8022007  3.2771133 -1.991029e-15
#> 
#> $var_explained
#>  PC1  PC2  PC3  PC4 
#> 0.55 0.28 0.16 0.00 
#>