estimateCellComposition: Estimate cell composition using reference

In sesame: Tools For Analyzing Illumina Infinium DNA Methylation Arrays

Description

This is a reference-based cell composition estimation. The function takes a reference methylation status matrix (rows for probes and columns for cell types, can be obtained by getRefSet function) and a query beta value measurement. The length of the target beta values should be the same as the number of rows of the reference matrix. The method assumes one unknown component. It outputs a list containing the estimated cell fraction, the error of optimization and methylation status of the unknown component.

Usage

estimateCellComposition(g, q, refine = TRUE, dichotomize = FALSE, ...)

Arguments

g	reference methylation
q	target measurement: length(q) == nrow(g)
refine	to refine estimate, takes longer
dichotomize	to dichotomize query beta value before estimate, this relieves unclean background subtraction
...	extra parameters for optimization, this includes temp - annealing temperature (0.5) maxIter - maximum iteration to stop after converge (1000) delta - delta score to reset counter (0.0001) verbose - output debug info (FALSE)

Value

a list of fraction, min error and unknown component methylation state

sesame documentation built on Nov. 15, 2020, 2:08 a.m.