kcsmooth: Smooth samples with KCsmart algorithm
In arnijohnsen/arjtools: Collection of tools used in copy number variation analysis

Description Usage Arguments Value

Wrapper to smooth aCGH data with algorithm from KCsmart.

1 2	kcsmooth(dat, mirrorLocs, sigma = 1e+06, sampleDensity = 50000, maxmem = 1000, verbose = T, what = "both", groups = NULL)

`dat`	data.table with aCGH data. First two colums should be names chrom and pos and contain information about the location of input probes. Chromosomes should be names "1", "Y", etc.
`mirrorLocs`	List containing the chromosome start, centromere and end positions
`sigma`	The kernel width
`sampleDensity`	The sample point matrix resolution
`maxmem`	This parameter controls memory usage, set to lower value to lower memory consumption
`verbose`	If set to false, no progress information is displayed
`what`	character, determines what data should be returned. "pos" for only positive KCscore, "neg" for only negative KCscore, "both" for both positive and negative KCscore.
`groups`	Numeric vector. If NULL is passed (default), each sample is smoothed seperately. If a numeric vector is passed, it indicates which samples should be grouped together before smoothing. For example, groups = c(1,1,2) indicates that the first two samples will be grouped before smothing and the third will be smoothed seperately. If a non-NULL value is passed, the return data frame will have columns equal to the number of groups passed. (This is the same as passing group = c(1,2,3,...)).

data.table with smoothed values. First column is chromosome, second column indicates if value is from positive or negative smoothed values. All other columns are smoothed values for each sample or group.

arnijohnsen/arjtools documentation built on May 10, 2019, 1:45 p.m.