% Generated by roxygen2: do not edit by hand
% Please edit documentation in R/metabarcoding_threshold.R
\docType{methods}
\name{threshold,metabarcoding.data-method}
\alias{threshold,metabarcoding.data-method}
\alias{threshold-methods,metabarcoding.data}
\title{Compute the cumulative threshold of read aboundances.}
\usage{
\S4method{threshold}{metabarcoding.data}(data, MARGIN = "sample",
  threshold = 0.97)
}
\arguments{
\item{data}{The \code{\linkS4class{metabarcoding.data}} instance
on normalisation have to be computed.}

\item{MARGIN}{Indicates if the sums have to be computed across 
samples or motus. 
Allowed values are :
\itemize{
  \item{'sample' or 1} for computing sum across samples
  \item{'motu' or 2} for computing sum across motus
  }}

\item{threshold}{a numeric value between 0 and 1 indicating which part of 
the signal must be conserved. Default value is setup to
0.97 (97% of the total signal).}
}
\value{
a numeric vector containing the limit aboundancy to consider for
        each sample or each MOTU according to the value of the \code{MARGIN} 
        parameter.
}
\description{
The method \code{threshold} of the class \code{\linkS4class{metabarcoding.data}}
computes the thresold to be used for conserving just a part of the global
signal. This thresold is computed by ranking aboundances by decreasing order.
The cululative sums of these ranked abondencies are computed and the aboundance
corresponding to the first sum greater than the threshold is returned as result.
}
\examples{
# load termite data set from the ROBITools sample data
data(termes)

# computes threshold value to used for keep 95\% of 
# the reads per MOTU

t = threshold(termes,MARGIN='motu',threshold=0.95)
 
}
\seealso{
\code{\linkS4class{metabarcoding.data}}, \code{\link{threshold.mask}}
}
\author{
Aurelie Bonin
}