1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
|
\name{remove.local.tag.anomalies}
\alias{remove.local.tag.anomalies}
%- Also NEED an '\alias' for EACH other topic documented here.
\title{ Restrict or remove positions with too many tags relative to
local background. }
\description{
In Solexa ChIP-seq experiments some anomalous positions contain
extremely high number of tags at the exact coordinates. The function
scans the chromosomes, determining local tag density based on a
provided \code{window.size}, doing two types of corrections:
1. removing all tags from positions that exceed local density by
\code{eliminate.fold}; 2. reducing the tag count at positions
exceeding \code{cap.fold} to the maximal allowed count. The
statistical significance of counts exceeding either of these two
threshold densities is calculated based on Poisson model, with
confidence interval determined by the \code{z.threshold} Z-score parameter.
}
\usage{
remove.local.tag.anomalies(tags,
window.size = 200,
eliminate.fold = 10,
cap.fold = 4,
z.threshold = 3)
}
%- maybe also 'usage' for other objects documented here.
\arguments{
\item{tags}{ Chromosome-list of tag vectors }
\item{window.size}{ Size of the window used to assess local
density. Increasing the window size considerably beyond the size of
the binding features will result in flattened profiles, with bound
positions exhibiting a difference of just 1 tag beyond the background. }
\item{eliminate.fold}{ Threshold definining fold-over background
density above which the position is considered anomalous and removed
completely.}
\item{cap.fold}{ Threshold fold-over background density above which
the position is capped to the maximum statistically likely given
local tag density }
\item{z.threshold}{ Z-score used to assess significance of a given
position exceeding either of the two density thresholds. }
%\item{var.base}{ Minimal level of tag count variance. This is used to
% guard against cases where variance of provided tags is near 0, as in
%the case of datasets filtered to unclude only unique reads. Defaults
% at 0.1}
}
\value{
A modified chromosome-wise tag vector list.
}
\references{ ~put references to the literature/web site here ~ }
\note{ ~~further notes~~
Increasing window.size to very large values will result in flat
profiles similar to those described by Zhang et al. "Model-based
Analysis of ChIP-Seq (MACS)." Genome Biol. 2008 Sep 17;9(9):R137.
}
|
