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
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
|
\name{getClans}
\alias{getClans}
\alias{getClips}
\alias{getSlices}
\alias{getDiversity}
\alias{diversity}
\title{
Clans, slices and clips
}
\description{
Functions for clanistics to compute clans, slices, clips for unrooted trees and functions to quantify the fragmentation of trees.
}
\usage{
getClans(tree)
getClips(tree, all=TRUE)
getSlices(tree)
getDiversity(tree, x, norm=TRUE, var.names = NULL, labels="new")
diversity(tree, X)
}
%- maybe also 'usage' for other objects documented here.
\arguments{
\item{tree}{An object of class phylo or multiPhylo (getDiversity).}
\item{all}{A logical, return all or just the largest clip. }
\item{x}{An object of class phyDat. }
\item{norm}{A logical, return Equitability Index (default) or Shannon Diversity. }
\item{var.names}{A vector of variable names. }
\item{labels}{see details. }
\item{X}{a data.frame}
}
\details{
Every split in an unrooted tree defines two complementary clans. Thus
for an unrooted binary tree with \eqn{n} leaves there are \eqn{2n - 3} edges,
and therefore \eqn{4n - 6} clans (including \eqn{n} trivial clans containing only
one leave).
Slices are defined by a pair of splits or tripartitions, which are not
clans. The number of distinguishable slices for a binary tree with
\eqn{n} tips is \eqn{2n^2 - 10n + 12}.
%A clip is a different type of partition as it is defined by evolutionary or cophenetic distance and not by the topology. Namely clips are groups of leaves for which the maximum pairwise distance is smaller than threshold.
%For a better separation we additionally demand that the maximum pairwise distance within a clip is lower than the distance between any member of the clip and any other tip.
A clip is a different type of partition, defining groups of leaves that are related in terms of evolutionary distances and not only topology.
Namely, clips are groups of leaves for which all pairwise path-length distances are smaller than a given threshold value (Lapointe et al. 2010).
There exists different numbers of clips for different thresholds, the largest (and trivial) one being the whole tree.
There is always a clip containing only the two leaves with the smallest pairwise distance.
Clans, slices and clips can be used to characterize how well a vector of
categorial characters (natives/intruders) fit on a tree. We will follow the definitions of Lapointe et al.(2010).
A complete clan is a clan that contains all leaves of a given state/color, but can
also contain leaves of another state/color. A clan is homogeneous if it
only contains leaves of one state/color.
\code{getDiversity} computes either the \cr
Shannon Diversity: \eqn{H = -\sum_{i=1}^{k}(N_i/N) log(N_i/N), N=\sum_{i=1}^{k} N_i}{H = -sum(N_i/N) * log(N_i/N), N=sum(N_i)} \cr
or the \cr
Equitability Index: \eqn{E = H / log(N)} \cr
where \eqn{N_i} are the sizes of the \eqn{k} largest homogeneous clans of intruders.
If the categories of the data can be separated by an edge of the tree then the E-value will be zero,
and maximum equitability (E=1) is reached if all intruders are in separate clans.
getDiversity computes these Intruder indices for the whole tree, complete clans and complete slices.
Additionally the parsimony scores (p-scores) are reported. The p-score indicates if the leaves contain only one color (p-score=0), if
the the leaves can be separated by a single split (perfect clan, p-score=1) or by a pair of splits (perfect slice, p-score=2).
So far only 2 states are supported (native, intruder), however it is also possible to recode several states
into the native or intruder state using contrasts, for details see section 2 in vignette("phangorn-specials").
Furthermore unknown character states are coded as ambiguous character, which can act either as native or intruder
minimizing the number of clans or changes (in parsimony analysis) needed to describe a tree for given data.
Set attribute labels to "old" for analysis as in Schliep et al. (2010) or to "new" for names which are more intuitive.
\code{diversity} returns a data.frame with the parsimony score for each tree and each levels of the variables in \code{X}. \code{X} has to be a \code{data.frame} where each column is a factor and the rownames of \code{X} correspond to the tips of the trees.
%TODO See also vignette("Clanistic").
}
\value{
getClans, getSlices and getClips return a matrix of partitions, a matrix of ones and zeros where rows correspond to
a clan, slice or clip and columns to tips. A one indicates that a tip belongs to a certain partition. \cr
getDiversity returns a list with tree object, the first is a data.frame of the equitability index or Shannon divergence and parsimony scores (p-score)
for all trees and variables. The data.frame has two attributes, the first is a splits object to identify the taxa of each tree and the second
is a splits object containing all partitions that perfectly fit.
}
\references{
Lapointe, F.-J., Lopez, P., Boucher, Y., Koenig, J., Bapteste, E. (2010) Clanistics: a multi-level perspective for harvesting unrooted gene trees. \emph{Trends in Microbiology} 18: 341-347
Wilkinson, M., McInerney, J.O., Hirt, R.P., Foster, P.G., Embley, T.M. (2007) Of clades and clans: terms for phylogenetic relationships in unrooted trees. \emph{Trends in Ecology and Evolution} 22: 114-115
Schliep, K., Lopez, P., Lapointe F.-J., Bapteste E. (2011) Harvesting Evolutionary Signals in a Forest of Prokaryotic Gene Trees, \emph{Molecular Biology and Evolution} 28(4): 1393-1405
}
\author{
Klaus Schliep \email{klaus.schliep@snv.jussieu.fr}
Francois-Joseph Lapointe \email{francois-joseph.lapointe@umontreal.ca}
}
\seealso{
\code{\link{parsimony}}, Consistency index \code{\link{CI}}, Retention index \code{\link{RI}}, \code{\link{phyDat}}
}
\examples{
set.seed(111)
tree = rtree(10)
getClans(tree)
getClips(tree, all=TRUE)
getSlices(tree)
set.seed(123)
trees = rmtree(10, 20)
X = matrix(sample(c("red", "blue", "violet"), 100, TRUE, c(.5,.4, .1)), ncol=5,
dimnames=list(paste('t',1:20, sep=""), paste('Var',1:5, sep="_")))
x = phyDat(X, type = "USER", levels = c("red", "blue"), ambiguity="violet")
plot(trees[[1]], "u", tip.color = X[trees[[1]]$tip,1]) # intruders are blue
(divTab <- getDiversity(trees, x, var.names=colnames(X)))
summary(divTab)
}
\keyword{ cluster }
|
