| Title: | alternative CDF environments (aka probeset mappings) |
|---|---|
| Description: | Convenience data structures and functions to handle cdfenvs |
| Authors: | Laurent Gautier <[email protected]> |
| Maintainer: | Laurent Gautier <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 2.69.0 |
| Built: | 2024-10-30 03:28:50 UTC |
| Source: | https://github.com/bioc/altcdfenvs |
Store the results of a call to matchAffyProbes.
Objects can be created by calls of the form new("AffyProbesMatch", ...).
An object will store the result of matching probe sequences against target sequences.
pm:Object of class "list":
each element is vector of index values
mm:Object of class "list":
each element is vector of index values
labels:Object of class "character"
chip_type:Object of class "character"
and of length 1.
probes:Object of class "ANY":
the probetable object used to perform the matches.
signature(x = "AffyProbesMatch", y =
"AffyProbesMatch"): combine two instances. This is can
be useful when splitting the list of target sequences to
parallelized the job.
signature(x = "AffyProbesMatch"):Show the
instance.
signature(object = "AffyProbesMatch"):
build an Hypergraph from the matches.
showClass("AffyProbesMatch")showClass("AffyProbesMatch")
append probe sets to a CdfEnvAffy
appendCdfEnvAffy(acdfenv, id, i, nocopy = TRUE)appendCdfEnvAffy(acdfenv, id, i, nocopy = TRUE)
acdfenv |
instance of class |
id |
identifier for the probe set to add |
i |
a |
nocopy |
whether to make a copy of the environment or not (see details) |
The matrix i must have one column per probe type. For
typical Affymetrix chip types, there are two probe types: "pm"
and "mm".
nocopy set to TRUE means that the environment is added
the probe set 'in-situ' (this can boost execution speed if you add a
lot of probe sets).
An CdfEnvAffy is returned
data(cdfenvEx) ## pm and mm probe set m <- matrix(1:10, ncol = 2) colnames(m) <- c("pm", "mm") appendCdfEnvAffy(cdfenvEx, "blabla", m) indexProbes(cdfenvEx, c("pm", "mm"), "blabla") ## pm only probe set m <- matrix(6:9, ncol = 1) colnames(m) <- c("pm") appendCdfEnvAffy(cdfenvEx, "blabla2", m) ## note that the unspecified "mm" were set to NA indexProbes(cdfenvEx, c("pm", "mm"), "blabla2")data(cdfenvEx) ## pm and mm probe set m <- matrix(1:10, ncol = 2) colnames(m) <- c("pm", "mm") appendCdfEnvAffy(cdfenvEx, "blabla", m) indexProbes(cdfenvEx, c("pm", "mm"), "blabla") ## pm only probe set m <- matrix(6:9, ncol = 1) colnames(m) <- c("pm") appendCdfEnvAffy(cdfenvEx, "blabla2", m) ## note that the unspecified "mm" were set to NA indexProbes(cdfenvEx, c("pm", "mm"), "blabla2")
Build CDF environment from Biostrings matchPDict results
buildCdfEnv.biostrings(apm, abatch = NULL, nrow.chip = NULL, ncol.chip = NULL, simplify = TRUE, x.colname = "x", y.colname = "y", verbose = FALSE)buildCdfEnv.biostrings(apm, abatch = NULL, nrow.chip = NULL, ncol.chip = NULL, simplify = TRUE, x.colname = "x", y.colname = "y", verbose = FALSE)
apm |
|
abatch |
|
nrow.chip |
number of rows for the chip type (see details) |
ncol.chip |
number of columns for the chip type (see details) |
simplify |
simplify the environment built (removing target names when there is no matching probe) |
x.colname |
column name |
y.colname |
column name |
verbose |
verbose |
Whenever an abatch is specified, nrow.chip and
ncol.chip are not needed. Specifying the an AffyBatch
in abatch is the easiest way to specify information about the
geometry of a chip type.
An instance of class CdfEnvAffy.
A class to hold the information necessary to handle the grouping of probes in set of probes, and to find XY coordinates of probes on a chip
Objects can be created by calls of the form new("CdfEnvAffy", ...).
Typically, there is an instance of the class for each type of chip
(e.g. Hu6800, HG-U95A, etc...).
envir:Object of class "environment". It has to
be thought of as a hashtable: the keys are probe set identifiers,
or gene names, and the values are indexes.
envName:Object of class "character". A name
for the environment.
index2xy:Object of class "function". The
function used to resolve index into xy coordinates. Unless you are
an advanced user, you probably want to ignore this (and rely on the
default provided with the package).
xy2index:Object of class "function". The
function used to resolve xy coordinates into index. Unless you are
an advanced user, you probably want to ignore this (and rely on the
default provided with the package).
nrow:Object of class "integer". The number of
rows of probes for the chip type.
ncol:Object of class "integer". The number of
columns of probes for the chip type.
probeTypes:Object of class "character". The
different types of probes stored for each probe set. In the case
of Affymetrix chips, the probes are typically perfect match
(pm) probes or mismatch probes (mm).
chipType:Object of class "character". The name
of the chip type the instance is associated with. This is useful
when one starts to create alternative mappings of the probes on a
chip (see associated vignette).
signature(object = "CdfEnvAffy",
i = "character", j = "missing", drop = "boolean"): subset a cdf, that
is return a new cdf containing only a subset of the probe
sets. The subset of probe sets to take is identified as a vector
of identifiers (mode "character").
signature(object = "CdfEnvAffy", "environment"): coerce an
instance of the class to an environment.
signature(object = "CdfEnvAffy", "Cdf"): coerce an
instance of the class to a Cdf.
signature(object="CdfEnvAffy"): Return the
names of the known probe sets (of course, it depends on the associated CDF).
signature(object = "CdfEnvAffy", i="integer"): convert
index values into XY coordinates.
signature(object = "CdfEnvAffy", which =
"character", probeSetNames = NULL): obtain the indexes for the probes associated wit
the probe set name probeSetNames. When probeSetNames
is set to NULL (default), the indexes are returned for the
probe sets defined on the chip. See indexProbes.CdfEnvAffy
signature(x = "CdfEnvAffy", y = "missing"): Plot
the chip. It mainly sets coordinates for further plotting (see
examples). See plot.CdfEnvAffy
signature(object = "CdfEnvAffy"): Print method.
signature(object = "CdfEnvAffy", x="integer", y="integer"): convert XY
coordinates into index values.
signature(object = "CdfEnvAffy"): convert XY
coordinates into index values.
Laurent Gautier
indexProbes.CdfEnvAffy, plot.CdfEnvAffy
## build an instance library(hgu95acdf) cdfenv.hgu95a <- wrapCdfEnvAffy(hgu95acdf, 640, 640, "HG-U95A") show(cdfenv.hgu95a) ## find the indexes for a probe set (pm only) ip <- indexProbes(cdfenv.hgu95a, "pm", "1000_at")[[1]] ## get the XY coordinates for the probe set xy <- index2xy(cdfenv.hgu95a, ip) ## plot the chip plot(cdfenv.hgu95a) ## plot the coordinates plotLocation(xy) ## subset the environment cdfenv.hgu95a.mini <- cdfenv.hgu95a["1000_at"]## build an instance library(hgu95acdf) cdfenv.hgu95a <- wrapCdfEnvAffy(hgu95acdf, 640, 640, "HG-U95A") show(cdfenv.hgu95a) ## find the indexes for a probe set (pm only) ip <- indexProbes(cdfenv.hgu95a, "pm", "1000_at")[[1]] ## get the XY coordinates for the probe set xy <- index2xy(cdfenv.hgu95a, ip) ## plot the chip plot(cdfenv.hgu95a) ## plot the coordinates plotLocation(xy) ## subset the environment cdfenv.hgu95a.mini <- cdfenv.hgu95a["1000_at"]
An example of CdfEnvAffy
data(cdfenvEx)data(cdfenvEx)
The format is: Formal class 'CdfEnvAffy' [package "altcdfenvs"] with 8 slots ..@ index2xy :function (object, i) ..@ xy2index :function (object, x, y) ..@ envir :length 2 <environment> ..@ envName : chr "ZG-DU33" ..@ nrow : int 100 ..@ ncol : int 100 ..@ probeTypes: chr [1:2] "pm" "mm" ..@ chipType : chr "ZG-DU33"
data(cdfenvEx) print(cdfenvEx)data(cdfenvEx) print(cdfenvEx)
A set of functions to handle cdfenvs
wrapCdfEnvAffy(cdfenv, nrow.chip, ncol.chip, chiptype, check = TRUE, verbose = FALSE) getCdfEnvAffy(abatch) buildCdfEnv.matchprobes(matches, ids, probes.pack, abatch=NULL, nrow.chip=NULL, ncol.chip=NULL, chiptype=NULL, mm=NA, simplify = TRUE, x.colname = "x", y.colname = "y", verbose=FALSE)wrapCdfEnvAffy(cdfenv, nrow.chip, ncol.chip, chiptype, check = TRUE, verbose = FALSE) getCdfEnvAffy(abatch) buildCdfEnv.matchprobes(matches, ids, probes.pack, abatch=NULL, nrow.chip=NULL, ncol.chip=NULL, chiptype=NULL, mm=NA, simplify = TRUE, x.colname = "x", y.colname = "y", verbose=FALSE)
abatch |
an |
cdfenv |
A cdfenv environment |
check |
perform consistency check or not |
chiptype |
A name for the chip type |
ids |
a vector of probe set identifiers for the matches |
matches |
a list as returned by the function
|
mm |
The value to store for MMs |
ncol.chip |
The number of columns for the chip type |
nrow.chip |
The number of rows for the chip type |
probes.pack |
The name of the probe package |
simplify |
Simplify the environment created by removing the ids without any matching probe |
x.colname, y.colname
|
see the |
verbose |
verbosity ( |
An instance of class CdfEnvAffy.
## See the main vignette## See the main vignette
make a copy of a CdfEnvAffy
copyCdfEnvAffy(acdfenv)copyCdfEnvAffy(acdfenv)
acdfenv |
instance of class |
Make a copy can be needed since a CdfEnvAffy contains an environment
A CdfEnvAffy
This function counts the number of times the probes in a CdfEnvAffy are found in this object.
countduplicated(x, incomparables = FALSE, verbose = FALSE)countduplicated(x, incomparables = FALSE, verbose = FALSE)
x |
An instance of |
incomparables |
(not implemented yet, keep away) |
verbose |
verbose or not |
An environment is returned. Each element in this
environment has the same identifier than its corresponding
probe set in the CdfEnvAffy-class and contains the number of
times a probe is in use in the environment (instead of an index number
in the CdfEnvAffy-class).
Laurent
get the names of the probe sets known to the CdfEnv
geneNames.CdfEnvAffy(object)geneNames.CdfEnvAffy(object)
object |
|
a vector of mode character
A function to get the XY coordinates from a probes sequences data.frame
getxy.probeseq(ppset.id = NULL, probeseq = NULL, i.row = NULL, xy.offset = NULL, x.colname = "x", y.colname = "y")getxy.probeseq(ppset.id = NULL, probeseq = NULL, i.row = NULL, xy.offset = NULL, x.colname = "x", y.colname = "y")
ppset.id |
The probe sets of interest (a vector of mode |
probeseq |
The probe sequence |
i.row |
Row indexes in the |
xy.offset |
Offset for the xy coordinates. if |
x.colname, y.colname
|
The probe sequence packages have seen the
names for the columns in their |
The data.frame passed as argument probeseq is expected
to have (at least) the following columns: Probe.X,
Probe.Y and Probe.Set.Name. When the argument
ppset.id is not null, the probe sets
A matrix of two columns. The first column contains x coordinates,
while the second column contains y coordinates.
The parameter xy.offset.one is here for historical
reasons. This should not be touched, the option in the affy
package should be modified if one wishes to modify this.
This function should not be confused with the methods index2xy
and similar. Here the the XY coordinate come from a data.frame
that stores information about an arbitrary number probes on the
chip. (See the ‘probe sequence’ data packages on Bioconductor, and the
package Biostrings).
The methods index2xy are meant to interact with instances of
class AffyBatch.
Laurent
##---- Should be DIRECTLY executable !! ----##---- Should be DIRECTLY executable !! ----
Functions to shuttle from indexes to XY coordinates.
index2xy(object, ...) xy2index(object, ...) index2xy.CdfEnvAffy(object, i) xy2index.CdfEnvAffy(object, x, y)index2xy(object, ...) xy2index(object, ...) index2xy.CdfEnvAffy(object, i) xy2index.CdfEnvAffy(object, x, y)
object |
An object of class |
i |
A vector of indexes. |
x, y
|
Vectors of X and Y coordinates. |
... |
Optional parameters (not used). |
A vector of integers (for xy2index methods), or a matrix of two
columns (for index2xy methods).
## To be done...## To be done...
A function to get the index for probes
indexProbes.CdfEnvAffy(object, which, probeSetNames = NULL)indexProbes.CdfEnvAffy(object, which, probeSetNames = NULL)
object |
|
which |
which kind of probe are of interest (see details). |
probeSetNames |
names of the probe sets of interest. If
|
The parameter which let one specify which category of probes
are of interest. In the case of Affymetrix chips, probes can be "pm"
probes or "mm" probes. It the parameter is set to c("pm",
"mm"), both are returned. Should other categories be defined, they
can be handled as well.
A list of indexes.
CdfEnvAffy-class, AffyBatch-class
Match the individual probes on an Affymetrix array to arbitrary targets.
mmProbes(probes) matchAffyProbes(probes, targets, chip_type, matchmm = TRUE, selectMatches = function(x) which(elementNROWS(x) > 0), ...)mmProbes(probes) matchAffyProbes(probes, targets, chip_type, matchmm = TRUE, selectMatches = function(x) which(elementNROWS(x) > 0), ...)
probes |
a |
targets |
a vector of references |
chip_type |
a name for the chip type. |
matchmm |
whether to match MM probes or not |
selectMatches |
a function to select matches (see Details). |
... |
further arguments to be passed to |
The matching is performed by the function
matchPDict. The man page
for that function will indicate what are the options it accepts.
In the case where a large number targets are given, like when
each target represents a possible mRNA, is it expected to have a
largely sparse incidence matrix, that is a low number of probes
matching every target. For that reason, only the index of matching
probes are associated with each given target, with the function
selectMatches giving the definition of what are
matching probes. The default function just count anything matching,
but the user can specify a more stringent definition if wanted.
mmProbes returns a vector of MM probe sequences.
matchAffyProbes returns an instance of AffyProbesMatch-class.
Laurent Gautier
matchPDict for
details on how the matching is performed, AffyProbesMatch-class
and buildCdfEnv.biostrings
library(hgu133aprobe) filename <- system.file("exampleData", "sample.fasta", package="altcdfenvs") fasta.seq <- readDNAStringSet(filename) targets <- as.character(fasta.seq) names(targets) <- sub("^>.+\\|(NM[^ \\|]+|Hs[^ \\|]+)\\| ? .+$", "", names(targets)) m <- matchAffyProbes(hgu133aprobe, targets, "HG-U133A")library(hgu133aprobe) filename <- system.file("exampleData", "sample.fasta", package="altcdfenvs") fasta.seq <- readDNAStringSet(filename) targets <- as.character(fasta.seq) names(targets) <- sub("^>.+\\|(NM[^ \\|]+|Hs[^ \\|]+)\\| ? .+$", "", names(targets)) m <- matchAffyProbes(hgu133aprobe, targets, "HG-U133A")
A function to set the axis and plot the outline for a CdfEnvAffy
## S3 method for class 'CdfEnvAffy' plot(x, xlab = "", ylab = "", main = x@chipType, ...)## S3 method for class 'CdfEnvAffy' plot(x, xlab = "", ylab = "", main = x@chipType, ...)
x |
a |
xlab |
label for the rows |
ylab |
label for the columns |
main |
label for the plot. The chip-type by default. |
... |
optional parameters to be passed to the underlying
function |
This function does not ‘plot’ much, but sets the coordinates for further plotting (see the examples).
Laurent
## See "CdfEnvAffy-class"## See "CdfEnvAffy-class"
Set of function to work with biological sequences stored in FASTA format.
countskip.FASTA.entries(con, linebreaks = 3000) grep.FASTA.entry(pattern, con, ...) ## S3 method for class 'FASTA' print(x, ...) read.FASTA.entry(con, linebreaks = 3000) read.n.FASTA.entries(con, n, linebreaks = 3000) read.n.FASTA.entries.split(con, n, linebreaks = 3000) read.n.FASTA.headers(con, n, linebreaks = 3000) read.n.FASTA.sequences(con, n, linebreaks = 3000) skip.FASTA.entry(con, skip, linebreaks = 3000) write.FASTA(x, file="data.fasta", append = FALSE)countskip.FASTA.entries(con, linebreaks = 3000) grep.FASTA.entry(pattern, con, ...) ## S3 method for class 'FASTA' print(x, ...) read.FASTA.entry(con, linebreaks = 3000) read.n.FASTA.entries(con, n, linebreaks = 3000) read.n.FASTA.entries.split(con, n, linebreaks = 3000) read.n.FASTA.headers(con, n, linebreaks = 3000) read.n.FASTA.sequences(con, n, linebreaks = 3000) skip.FASTA.entry(con, skip, linebreaks = 3000) write.FASTA(x, file="data.fasta", append = FALSE)
append |
append to the file (or not) |
con |
|
file |
a file name |
linebreaks |
(to optimize the parsing, probably safe to leave it as it is) |
n |
number of entries to read |
pattern |
a pattern (to be passed to the function |
skip |
number of entries to skip |
x |
a FASTA sequence object |
... |
optional arguments to be forwarded to the function
|
countskip.FASTA.entries skips the remaining FASTA entries
currently remaining in the connection and return the count.
grep.FASTA.entry returns the next FASTA entry in the connection
that matches a given regular expression.
print.FASTA prints a FASTA object.
read.FASTA.entry reads the next FASTA entry in the connection.
read.n.FASTA.entries reads the n next FASTA entries and
returns a list of FASTA objects.
read.n.FASTA.entries.split reads the n next FASTA
entries and returns a list of two elements: headers and sequences.
read.n.FASTA.headers reads the n next FASTA headers.
read.n.FASTA.sequences reads the n next FASTA sequences.
skip.FASTA.entry skips a given number of FASTA entries.
write.FASTA write a FASTA object into a connection.
The value returned depends on the function. See above.
Laurent Gautier
filename <- system.file("exampleData", "sample.fasta", package="altcdfenvs") con <- file(filename, open="r") fasta.seq <- grep.FASTA.entry("NM_001544\\.2", con) close(con) print(fasta.seq)filename <- system.file("exampleData", "sample.fasta", package="altcdfenvs") con <- file(filename, open="r") fasta.seq <- grep.FASTA.entry("NM_001544\\.2", con) close(con) print(fasta.seq)
A function to remove probes in an environment, given their index.
removeIndex(x, i, simplify = TRUE, verbose = FALSE)removeIndex(x, i, simplify = TRUE, verbose = FALSE)
x |
An instance of |
i |
A vector of indexes (integers !). |
simplify |
Simply the resulting |
verbose |
verbose output or not. |
The probes to be removed are set to NA in the CdfEnvAffy.
When simplify is set to TRUE the probe sets are
simplified whenever possible. For example, if both pm and mm for the
same probe pair are set to NA, then the probe pair is removed
from the probe set.
An instance of CdfEnvAffy-class is returned.
Laurent Gautier
## use plasmodiumanopheles chip as an example if (require(plasmodiumanophelescdf)) { ## wrap in a (convenient) CdfEnvAffy object planocdf <- wrapCdfEnvAffy(plasmodiumanophelescdf, 712, 712, "plasmodiumanophelescdf") print(planocdf) ## ask for the probe indexed '10759' to be removed ## (note: if one wishes to remove from X/Y coordinates, ## the function xy2index can be of help). planocdfCustom <- removeIndex(planocdf, as.integer(10759)) ## let see what happened (we made this example knowing in which ## probe set the probe indexed '10759' is found). indexProbes(planocdf, "pm", "200000_s_at") indexProbes(planocdfCustom, "pm", "200000_s_at") ## The 'second' pm probe (indexed '10579') in the probe set is now set ## to NA. }## use plasmodiumanopheles chip as an example if (require(plasmodiumanophelescdf)) { ## wrap in a (convenient) CdfEnvAffy object planocdf <- wrapCdfEnvAffy(plasmodiumanophelescdf, 712, 712, "plasmodiumanophelescdf") print(planocdf) ## ask for the probe indexed '10759' to be removed ## (note: if one wishes to remove from X/Y coordinates, ## the function xy2index can be of help). planocdfCustom <- removeIndex(planocdf, as.integer(10759)) ## let see what happened (we made this example knowing in which ## probe set the probe indexed '10759' is found). indexProbes(planocdf, "pm", "200000_s_at") indexProbes(planocdfCustom, "pm", "200000_s_at") ## The 'second' pm probe (indexed '10579') in the probe set is now set ## to NA. }
Transform to an hypergraph
toHypergraph(object, ...)toHypergraph(object, ...)
object |
Object derived from class |
... |
Unused. |
An Hypergraph-class object.
Remove duplicated elements from a CdfEnvAffy
## S3 method for class 'CdfEnvAffy' unique(x, incomparables = FALSE, simplify = TRUE, verbose = FALSE, ...)## S3 method for class 'CdfEnvAffy' unique(x, incomparables = FALSE, simplify = TRUE, verbose = FALSE, ...)
x |
An instance of |
incomparables |
(not yet implemented) |
simplify |
simplify the result |
verbose |
verbose or not |
... |
(here for compatibility with the generic |
The parameter simplify has the same function as the one with
the same name in countduplicated.
An instance of CdfEnvAffy-class in which probes used several
times are removed.
The function differs slightly from the generic
unique. Here the elements found in several place a merely removed.
Laurent
##not yet here...##not yet here...
Tries to see if a CdfEnvAffy, or a pair of AffyBatch / CdfEnvAffy is valid.
validAffyBatch(abatch, cdfenv) validCdfEnvAffy(cdfenv, verbose=TRUE) printValidCdfEnvAffy(x)validAffyBatch(abatch, cdfenv) validCdfEnvAffy(cdfenv, verbose=TRUE) printValidCdfEnvAffy(x)
abatch |
instance of |
cdfenv |
instance of |
verbose |
verbose or not |
x |
object returned by |
The function validAffyBatch calls in turn
validCdfEnvAffy.
AffyBatch-class, CdfEnvAffy-class
## To be done...## To be done...