[adegenet-commits] r622 - in pkg: R man
noreply at r-forge.r-project.org
noreply at r-forge.r-project.org
Mon May 10 18:01:24 CEST 2010
Author: jombart
Date: 2010-05-10 18:01:23 +0200 (Mon, 10 May 2010)
New Revision: 622
Modified:
pkg/R/seqTrack.R
pkg/man/seqTrack.Rd
Log:
seqTrack doc done except example. Code seems ok. Got to check haploSim before going further...
Modified: pkg/R/seqTrack.R
===================================================================
--- pkg/R/seqTrack.R 2010-05-10 15:47:18 UTC (rev 621)
+++ pkg/R/seqTrack.R 2010-05-10 16:01:23 UTC (rev 622)
@@ -298,28 +298,20 @@
###########################
## get.likelihood.seqTrack
###########################
-get.likelihood.seqTrack <-function(x, method=("genetic"), mu=NULL, seq.length=NULL,...){
- method <- match.arg(method)
- if(method=="genetic"){ # p(nb mutations occur in the time interval)
- if(any(na.omit(x$weight - round(x$weight)) > 1e-10)){
- warning("Non-integer weights: number of mutations expected in x$weight.")
- }
+get.likelihood.seqTrack <- function(x, mu, haplo.length,...){
+ if(any(na.omit(x$weight - round(x$weight)) > 1e-10)){
+ warning("Non-integer weights: number of mutations expected in x$weight.")
+ }
- if(is.null(mu)) stop("mu is required.")
- if(is.null(seq.length)) stop("seq.length is required.")
+ dates <- as.POSIXct(x$date)
+ anc.dates <- as.POSIXct(x$ances.date)
+ nb.days <- abs(as.integer(anc.dates-dates))
+ nb.mut <- x$weight
+ ##mu <- mu/365
+ ##mu <- mu*nb.days
- dates <- as.POSIXct(x$date)
- anc.dates <- as.POSIXct(x$ances.date)
- nb.days <- abs(as.integer(anc.dates-dates))
- nb.mut <- x$weight
- ##mu <- mu/365
- ##mu <- mu*nb.days
+ res <- dbinom(nb.mut, size=seq.length*nb.days, prob=mu)
- res <- dbinom(nb.mut, size=seq.length*nb.days, prob=mu)
- } else{
- cat("Method not implemented.")
- }
-
return(res)
} # end get.likelihood.seqTrack
@@ -448,12 +440,6 @@
## }
-
-
-
-
-
-
## ##################
## ## .pAbeforeBfast
## ##################
Modified: pkg/man/seqTrack.Rd
===================================================================
--- pkg/man/seqTrack.Rd 2010-05-10 15:47:18 UTC (rev 621)
+++ pkg/man/seqTrack.Rd 2010-05-10 16:01:23 UTC (rev 622)
@@ -1,16 +1,11 @@
\name{seqTrack}
\alias{seqTrack}
-% \alias{seqTrackG}
\alias{seqTrack.default}
-% \alias{seqTrackG.default}
+\alias{seqTrack.matrix}
\alias{plotSeqTrack}
-% \alias{optimize.seqTrack}
-% \alias{optimize.seqTrack.default}
\alias{get.likelihood}
\alias{get.likelihood.seqTrack}
\alias{seqTrack-class}
-% \alias{get.result.by}
-% \alias{get.consensus}
\alias{as,seqTrack,graphNEL-method}
\alias{coerce,seqTrack,graphNEL-method}
\title{SeqTrack algorithm for reconstructing genealogies}
@@ -31,19 +26,23 @@
\usage{
seqTrack(\ldots)
-\method{seqTrack}{matrix}(x, x.names, x.dates, best = c("min", "max"), prox.mat = NULL,
- mu = NULL, haplo.length = NULL, \dots)
+\method{seqTrack}{matrix}(x, x.names, x.dates, best = c("min", "max"),
+ prox.mat = NULL, mu = NULL, haplo.length = NULL, \dots)
-plotSeqTrack(x, xy, use.arrows=TRUE, annot=TRUE, labels=NULL,
- col=NULL, bg="grey", add=FALSE, quiet=FALSE, date.range=NULL,
- plot=TRUE, \dots)
+plotSeqTrack(x, xy, use.arrows=TRUE, annot=TRUE, labels=NULL, col=NULL,
+ bg="grey", add=FALSE, quiet=FALSE,
+ date.range=NULL, plot=TRUE, \dots)
+
+get.likelihood(\ldots)
+\method{get.likelihood}{seqTrack}(x, mu, haplo.length, \ldots)
}
\arguments{
\item{x}{for seqTrack, a matrix giving weights to pairs of ancestries
such that x[i,j] is the weight of 'i ancestor of j'. For
- plotSeqTrack, a seqTrack object.}
- \item{seq.names}{a character vector giving the labels of the haplotypes/genotypes}
- \item{seq.dates}{a vector of collection dates for the sampled
+ plotSeqTrack and get.likelihood. seqTrack, a \code{seqTrack}
+ object.}
+ \item{x.names}{a character vector giving the labels of the haplotypes/genotypes}
+ \item{x.dates}{a vector of collection dates for the sampled
haplotypes/genotypes. Dates must have the POSIXct format. See
\code{details} or \code{?as.POSIXct} for more information.}
\item{best}{a character string matching 'min' or 'max', indicating
@@ -93,21 +92,21 @@
outbreaks from genetic data: a graph approach. Heredity.
}
\value{
- === output of seqTrack ===
+ === output of seqTrack ===\cr
seqTrack function returns data.frame with the class \code{seqTrack},
in which each row is an inferred ancestry described by the following columns:
- - id: indices identifying haplotypes/genotypes
- - ances: index of the inferred ancestor
- - weight: weight of the inferred ancestries
- - date: date of the haplotype/genotype
+ - id: indices identifying haplotypes/genotypes\cr
+ - ances: index of the inferred ancestor\cr
+ - weight: weight of the inferred ancestries\cr
+ - date: date of the haplotype/genotype\cr
- ances.date: date of the ancestor
- === output of plotSeqTrack ===
+ === output of plotSeqTrack ===\cr
This graphical function invisibly returns the coordinates of the
arrows/segments drawn and their colors, as a data.frame.
}
\details{
- === Maximum parsimony genealogies ===
+ === Maximum parsimony genealogies ===\cr
Maximum parsimony genealogies can be obtained easily using this
implementation of seqTrack. One has to provide in \code{x} a matrix of
genetic distances. The most straightforward distance is the number of
@@ -120,21 +119,21 @@
resolving ties in equally parsimonious ancestors using maximum
likelihood.
- === Likelihood of observed genetic differentiation ===
+ === Likelihood of observed genetic differentiation ===\cr
The probability of oberving a given number of mutations between a
sequence and its ancestor can be computed using
- \code{\link{get.likelihood.seqTrack}}. Note that this is only possible
+ \code{get.likelihood.seqTrack}. Note that this is only possible
if \code{x} contained number of mutations.
- === Converting seqTrack objects to graphs ===
+ === Converting seqTrack objects to graphs ===\cr
seqTrack objects can be converted to \linkS4class{graphNEL} objects,
which can in turn be plotted and manipulated using classical graph
tools. Simply use 'as(x, "graphNEL")' where 'x' is a seqTrack
object. This functionality requires the \code{graph} package. Note
that this is to be installed from Bioconductor, likely using the following
- command lines:
- source("http://bioconductor.org/biocLite.R")
+ command lines:\cr
+ source("http://bioconductor.org/biocLite.R")\cr
biocLite("graph")
Also note that the R package Rgraphviz (also on Bioconductor) provides
@@ -144,6 +143,6 @@
\seealso{
\code{\link[ape]{dist.dna}} to compute pairwise genetic distances in aligned sequences.
}
-\example{
+\examples{
}
\ No newline at end of file
More information about the adegenet-commits
mailing list