Function to get a set of background peaks for each peak based on GC content and # of fragments across all samples
getBackgroundPeaks(object, ...) # S4 method for SummarizedExperiment getBackgroundPeaks(object, bias = rowData(object)$bias, niterations = 50, w = 0.1, bs = 50) # S4 method for RangedSummarizedExperiment getBackgroundPeaks(object, bias = rowRanges(object)$bias, niterations = 50, w = 0.1, bs = 50) # S4 method for MatrixOrmatrix getBackgroundPeaks(object, bias, niterations = 50, w = 0.1, bs = 50)
| object | fragment counts as SummarizedExperiment, RangedSummarized, Matrix, or matrix |
|---|---|
| ... | addtional arguments |
| bias | vector of values for some bias signal for each row of object |
| niterations | number of background peaks to sample |
| w | parameter controlling similarity of background peaks |
| bs | bin size parameter |
matrix with one row per peak and one column per iteration. values in a row represent indices of background peaks for the peak with that index
Background peaks are chosen by sampling peaks based on similarity in GC content and # of fragments across samples using the Mahalanobis distance. The w paramter controls how similar background peaks should be. The bs parameter controls the precision with which the similarity is computed; increasing bs will make the function run slower. Sensible default parameters are chosen for both.
SummarizedExperiment: method for SummarizedExperiment
RangedSummarizedExperiment: method for RangedSummarizedExperiment
MatrixOrmatrix: method for Matrix or matrix
# NOT RUN { # Load very small example counts (already filtered) data(mini_counts, package = "chromVAR") # get background peaks bgpeaks <- getBackgroundPeaks(mini_counts) # }