Description Usage Arguments Author(s) Examples
View source: R/MethylResolver.R
This function deconvolves bulk methylation data into cell type fractions given a reference signature matrix for each cell type expected in the bulk profile. It also allows for the estimation of unknown content (purity) which allows both relative and absolute cell type fractions to be reported.
1 2 3 4 |
methylMix |
MxN (CpGs by samples) matrix of bulk methylation sequencing. Required. |
methylSig |
MxW (CpGs by cell types) matrix of purified cell type methylation profiles. Default signature is the MethylResolver leukocyte signature. |
betaPrime |
Deconvolutions are calculated using methylation Beta' values (1-Beta). Indicate whether you are supplying Beta' values (TRUE), or Beta values (FALSE). Default is TRUE. |
outputPath |
Path to output directory. Default is "./". |
outputName |
Prefix name to give to the deconvolution output. Default is "MethylResolver". |
doPar |
Whether to use parallel processing to speed up the deconvolution computation if many samples are present. Default is FALSE. |
numCores |
Number of cores used for parallel processing to speed up the deconvolution of many samples. Requires doPar = TRUE. Default is 1. numCores = "auto" is max number of cores available minus one. |
alpha |
Set the alpha parameter for LTS deconvolution. This is the fraction of optimal CpGs from the signature matrix which are used for deconvolution. Must be between 0 and 1. Users can specify a vector or a single number. If a vector is specified, a grid search of the values is conducted and the alpha value that results in the lowest RMSE between the original and reconstructed mixture is selected. Default is seq(0.5,0.9,by = 0.05). |
absolute |
Whether to compute tumor purity and absolute cell type fractions. Default is TRUE. |
purityModel |
Random Forest model to predict mixture purity (unknown content) which allows the calculation of absolute cell type fractions. Required if absolute is TRUE. Default is our RF model trained on the consensus purity estimate (CPE) using TCGA data. |
Douglas Arneson
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 | # Deconvolution with default signature and calculating
# absolute fractions with default RF model:
MethylResolver(methylMix = MethylMix)
# Deconvolution with default signature and calculating
# absolute fractions with default RF model while
# specifying a particular alpha value:
MethylResolver(methylMix = MethylMix, alpha = 0.5)
# Deconvolution with default signature and only
# calculating relative fractions:
MethylResolver(methylMix = MethylMix, absolute = FALSE)
# Specify your own signature matrix and RF model for
# calculating absolute fractions:
MethylResolver(methylMix = MethylMix, methylSig =
MethylSig, purityModel = RFmodel)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.