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
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
|
\name{NChannelSet-class}
\docType{class}
\alias{NChannelSet-class}
\alias{NChannelSet}
\alias{class.NChannelSet}
\alias{assayData<-,NChannelSet,environment-method}
\alias{assayData<-,NChannelSet,list-method}
\alias{channelNames,NChannelSet-method}
\alias{channelNames<-,NChannelSet,character-method}
\alias{channelNames<-,NChannelSet,list-method}
\alias{channel,NChannelSet,character-method}
\alias{sampleNames,NChannelSet-method}
\alias{sampleNames<-,NChannelSet,list-method}
\alias{selectChannels,NChannelSet,character-method}
\alias{initialize,NChannelSet-method}
\title{Class to contain data from multiple channel array technologies}
\description{
Container for high-throughput assays and experimental meta-data. Data
are from experiments where a single \sQuote{chip} contains several
(more than 1) different \sQuote{channels}. All channels on a chip have
the same set of\sQuote{features}. An experiment consists of a
collection of several N-channel chips; each chip is a \sQuote{sample}.
An \code{NChannelSet} provides a way to coordinate assay data
(expression values) with phenotype information and references to chip
annotation data; it extends the \code{\link{eSet}} class.
An \code{NChannelSet} allows channels to be extracted (using the
\code{channels} method, mentioned below), and subsets of features or
samples to be selected (using \code{[<features>,
<samples>]}). Selection and subsetting occur so that relevant
phenotypic data is maintained by the selection or subset.}
\section{Objects from the Class}{
Objects can be created by calls of the form \code{NChannelSet(
assayData, phenoData, ...)}. See the examples below.
}
\section{Slots}{
\describe{
\item{\code{assayData}:}{
Object of class \code{\link{AssayData}}, usually
an environment containing matrices of identical size. Each matrix
represents a single channel. Columns in each matrix correspond to
samples, rows to features. Once created, \code{NChannelSet} manages
coordination of samples and channels.
}
\item{\code{phenoData}:}{
Object of class \code{\link{AnnotatedDataFrame}}.
The \code{data} component of the \code{AnnotatedDataFrame} is
\code{data.frame} with number of rows equal to the number of
samples. Columns of the \code{data} component correspond to measured
covariates.
The \code{varMetadata} component consists of mandatory columns
\code{labelDescription} (providing a textual description of each column
label in the \code{data} component) and \code{channel}. The \code{channel}
of \code{varMetadata} is a \code{factor}, with levels equal to the names
of the \code{assayData} channels, plus the special symbol
\code{_ALL_}. The \code{channel} column is used to indicate which
channel(s) the corresponding column in the \code{data} component
of \code{AnnotatedDataFrame} correspond; the \code{_ALL_} symbol
indicates that the \code{data} column is applicable to all
channels. \code{varMetadata} may contain additional columns with
arbitrary information.
Once created, \code{NChannelSet} coordinates selection and subsetting
of channels in \code{phenoData}.
}
\item{\code{featureData}:}{
Object of class \code{\link{AnnotatedDataFrame}}, used to contain
feature data that is unique to this experiment; feature-level
descriptions common to a particular chip are usually referenced
through the \code{annotation} slot.
}
\item{\code{experimentData}:}{
Object of class \code{\link{MIAME}} containing descriptions of the
experiment.
}
\item{\code{annotation}:}{
Object of class \code{"character"}. Usually a length-1 character
string identifying the chip technology used during the
experiment. The annotation string is used to retrieve information
about features, e.g., using the \code{annotation} package.}
\item{\code{protocolData}:}{
Object of class \code{"character"}. A character vector
identifying the dates the samples were scanned during the
experiment.}
\item{\code{.__classVersion__}:}{
Object of class \code{\link{Versions}}, containing
automatically created information about the class definition
Biobase package version, and other information about the user
system at the time the instance was created. See
\code{\link{classVersion}} and \code{\link{updateObject}} for
examples of use.
}
}
}
\section{Extends}{
Class \code{"\linkS4class{eSet}"}, directly.
Class \code{"\linkS4class{VersionedBiobase}"}, by class "eSet", distance 2.
Class \code{"\linkS4class{Versioned}"}, by class "eSet", distance 3.
}
\section{Methods}{
Methods with class-specific functionality:
\describe{
\item{\code{channel(object, name, ...)}}{
\code{signature(object="NChannelSet", name="character")}.
Return an \code{ExperessionSet} created from the channel and
corresponding phenotype of argument \code{name}. \code{name} must
have length 1. Arguments \code{...} are rarely used, but are
passed to the \code{ExpressionSet} constructor, for instance
to influence \code{storage.mode}.
}
\item{\code{channelNames(object)}, \code{channelNames(object) <- value}}{
\code{signature(object = "NChannelSet")}.
Obtain, reorder, or rename channels contained in
\code{object}. See \code{\link{channelNames}}.
}
\item{\code{selectChannels(object, names, ...}}{
\code{signature(object = "NChannelSet", names = "character")}.
Create a new \code{NChannelSet} from \code{object}, containing
only channels in \code{names}. The \code{...} is not used by this
method.
}
\item{\code{object[features, samples]}}{
\code{signature(object = "NChannelSet",
features = "ANY", samples = "ANY")}.
Create a new \code{NChannelSet} from \code{object}, containing
only elements matching \code{features} and \code{samples}; either
index may be missing, or a character, numeric, or logical vector.
}
\item{\code{sampleNames(object) <- value}}{
\code{signature(object = "NChannelSet", value = "list")} assign
each (named) element of \code{value} to the \code{sampleNames} of
the correspondingly named elements of \code{assayData} in
\code{object}.
}
}
Methods with functionality derived from \code{\link{eSet}}:
\code{annotation}, \code{annotation<-}, \code{assayData},
\code{assayData<-}, \code{classVersion}, \code{classVersion<-},
\code{dim}, \code{dims}, \code{experimentData}, \code{experimentData<-},
\code{featureData}, \code{featureData<-}, \code{phenoData},
\code{phenoData<-}, \code{protocolData}, \code{protocolData<-},
\code{pubMedIds}, \code{pubMedIds<-}, \code{sampleNames},
\code{sampleNames<-}, \code{storageMode}, \code{storageMode<-},
\code{varMetadata}, \code{varMetadata<-}, \code{isCurrent},
\code{isVersioned}, \code{updateObject}.
Additional methods: \code{coerce} (\sQuote{as}, to convert between
objects, if possible), \code{initialize} (used internally for
creating objects), \code{show} (invoked automatically when the
object is displayed to the screen)
}
\author{Martin Morgan <mtmorgan@fhcrc.org>}
\seealso{
\code{\linkS4class{eSet}}, \code{\linkS4class{ExpressionSet}}.
}
\examples{
## An empty NChannelSet
obj <- NChannelSet()
## An NChannelSet with two channels (R, G) and no phenotypic data
obj <- NChannelSet(R=matrix(0,10,5), G=matrix(0,10,5))
## An NChannelSet with two channels and channel-specific phenoData
R <- matrix(0, 10, 3, dimnames=list(NULL, LETTERS[1:3]))
G <- matrix(1, 10, 3, dimnames=list(NULL, LETTERS[1:3]))
assayData <- assayDataNew(R=R, G=G)
data <- data.frame(ChannelRData=numeric(ncol(R)),
ChannelGData=numeric(ncol(R)),
ChannelRAndG=numeric(ncol(R)))
varMetadata <- data.frame(labelDescription=c(
"R-specific phenoData",
"G-specific phenoData",
"Both channel phenoData"),
channel=factor(c("R", "G", "_ALL_")))
phenoData <- AnnotatedDataFrame(data=data, varMetadata=varMetadata)
obj <- NChannelSet(assayData=assayData, phenoData=phenoData)
obj
## G channel as NChannelSet
selectChannels(obj, "G")
## G channel as ExpressionSet
channel(obj, "G")
## Samples "A" and "C"
obj[,c("A", "C")]
}
\keyword{classes}
|
