getCluster: Detect clusters of genomic features (i.e TFBS) in a specified...
In fcScan: fcScan for detecting clusters of coordinates with user defined options

Description Usage Arguments Details Value Author(s) Examples

View source: R/getCluster.R

Given a data frame or a GRanges object or a list of bed/vcf files, this function will look for clusters of genomic features found within a user defined window size and satisfiying user defined categorical conditions.

1 2	getCluster(x, w, c, overlap = 0, greedy = FALSE, seqnames = NULL, s = "*", order = NULL, site_orientation = NULL, site_overlap = 0, verbose = FALSE)

`x`	Vector of file names or a data frame
`w`	Value for the desired cluster size (numeric)
`c`	A named vector for condition relative to each site name
`overlap`	numeric; If negative, it will correspond to the maximum overlap allowed between consecutive clusters, if positive, it will correspond to the minimum gap allowed between consecutive clusters. The default is set to 0
`greedy`	logical; If FALSE, the formed clusters will contain the exact combination of sites defined in "c". If TRUE, clusters containing the condition and more will be labeled as TRUE as long as window size is below or equal the defined value. The default is set to FALSE
`seqnames`	character; seqname/chromosome to cluster on. The default is set to NULL so the clustering will be performed on all seqnames
`s`	character; The strand to cluster on. Can be "+", "-" or "". The default is set to "" which indicate both clusters
`order`	vector; A vector containing the desired order of sites within the identified clusters. The default is set to NULL indicating that the order of sites is not important
`site_orientation`	vector; A vector containing the desired orientation of the sites within the identified clusters. The default is set to NULL
`site_overlap`	numeric; Same as "overlap" but for sites. If negative, it will correspond to the maximum overlap allowed between consecutive sites. If positive, it will correspond to the minimum gap allowed between consecutive sites. The default is set to 0
`verbose`	logical; A If FALSE, only clusters that passed combination condition are shown. used for debugging purposes. By default it is set to FALSE

The function getCluster will cluster coordinates based on a user defined number of sites and window size. The user needs to specify the condition for clustering, which is the sites and the number of sites required in each cluster. The user can exclude sites from clusters. this should be specified also in the condition for clustering. the user can also set the distance required between consective clusters using the overlap argument. If overlap is a negative, this will represent the maximum overlap allowed between clusters. If overlap is positive, this means that the clusters should have a minimum gap of the given value.

The user can also choose to cluster on a specific strand by using the strand option, and specific seqname by specifying the seqnames as an argument.

x is a vector of files or a data frame. When data frame is given as input, it should have 5 columns seqnames start end strand site. start and end column need to be numeric or integer and the rest of the columns are of type character.

The Greedy option controls if more sites are allowed in the cluster. When set to TRUE, getCluster will return clusters containing the required number of sites and more as long as the window size condition is satisfied. When set to FALSE, getCluster will return clusters containing the exact number of the required sites. Default is set to FALSE.

order allows the user to choose whether the clusters should contain sites ordered in a certain way. When set to NULL, sites order is not important. When defined, the order of the sites in the cluster must abide by the user defined order. Default is NULL.

The returned value is a GRanges object containing the clusters.

Abdullah El-Kurdi

Ghiwa Khalil

Georges Khazen

Pierre Khoueiry

x = data.frame(seqnames = rep("chr1", times = 16),
start = c(10,17,25,27,32,41,47,60,70,87,94,99,107,113,121,132),
end = c(15,20,30,35,40,48,55,68,75,93,100,105,113,120,130,135),
strand = rep("+", 16),
site = c("s1","s2","s2","s1","s2","s1","s1","s2",
"s1","s2","s2","s1","s2","s1","s1","s2"))

clusters = getCluster(x, w = 25, c = c("s1"=1,"s2"=2),
greedy = TRUE, overlap = -5, s = "+", order = c("s1","s2","s1"))