-
Notifications
You must be signed in to change notification settings - Fork 4
/
file.R
248 lines (216 loc) · 10.5 KB
/
file.R
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
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
load.file <- function(filename)
{
sampleRD <- NULL
file.extension <- strsplit(filename, split="\\.")[[1]]
file.type <- file.extension[length(file.extension)]
if(file.type=="cdf") sampleRD <- load.ncdf4(filename)
if(file.type=="CDF") sampleRD <- load.ncdf4(filename)
if(file.type=="mzXML" || file.type=="xml") sampleRD <- load.xml(filename)
if(file.type=="mzML") sampleRD <- load.xml(filename)
if(file.type=="rdata") sampleRD <- load.erahrd(filename)
if(is.null(sampleRD)) stop("File extension not recognized. Avalible extensions are: .cdf, .mzXML and .xml")
sampleRD
}
load.erahrd <- function(filename)
{
sampleRD <- 0
samp.file <- load(filename)
sampleObject <- sampleRD
remove(sampleRD)
sampleObject
}
load.xml <- function(filename)
{
if(requireNamespace("mzR", quietly = TRUE)) {
xmlO <- mzR::openMSfile(filename)
metadata <- mzR::runInfo(xmlO)
scans <- 1:metadata$scanCount
lowMZ <- metadata$lowMz
highMZ <- metadata$highMz
if(lowMZ==0 | highMZ==0 | scans[2]==0)
{
peakLst <- mzR::peaks(xmlO)
mzVct <- unlist(lapply(peakLst, function(x) x[,1]))
lowMZ <- min(mzVct, na.rm=T)
highMZ <- max(mzVct, na.rm=T)
scans <- which(unlist(lapply(peakLst, function(x) nrow(x)))!=0)
}
lowMZ <- round(lowMZ + 0.5)
highMZ <- round(highMZ + 0.5)
StartTime <- metadata$dStartTime
ScansPerSecond <- 1/((metadata$dEndTime - metadata$dStartTime)/metadata$scanCount)
log <- utils::capture.output(raw.data <- mzR::get3Dmap(object = xmlO, scans = scans, lowMz = lowMZ, highMz = highMZ, resMz = 1))
sampleRD <- new("RawDataParameters", data = raw.data, min.mz = lowMZ, max.mz = highMZ, start.time = StartTime, mz.resolution = 1, scans.per.second = ScansPerSecond)
return(sampleRD)
}
else {
msg <- c("mzR is not installed. eRah and Baitmet can operate withouth mzR, unless you want to process .mzXML files (as in this case). To install the mzR package and be able to use mzXML files, please visit its bioconductor website: http://bioconductor.org/packages/release/bioc/html/mzR.html\nOr, alternatively, execute the following R code:\n\t\t\n\t\t## try http:// if https:// URLs are not supported \n\t\tsource('https://bioconductor.org/biocLite.R')\n\t\tbiocLite('mzR')")
warning(msg)
}
}
load.ncdf4 <- function(filename)
{
if(!requireNamespace("ncdf4", quietly = TRUE)){
msg <- c("ncdf4 is not installed. eRah and Baitmet can operate withouth ncdf4, unless you want to process .CDF files (as in this case). To install the ncdf4 package use: install.packages('ncdf4')")
stop(msg)
}
isExact <- FALSE
measurement = ncdf4::nc_open(filename)
mass_values <- ncdf4::ncvar_get(measurement, "mass_values")
rndSmplColl <- sample(mass_values, 500)
if (any(rndSmplColl != (rndSmplColl^2/trunc(rndSmplColl)))) isExact <- TRUE
mass_intensities <- ncdf4::ncvar_get(measurement, "intensity_values")
scan_indexes <- ncdf4::ncvar_get(measurement, "scan_index")
min_mz <- round(min(mass_values)) - 1
max_mz <- round(max(mass_values)) + 1
start_time <- as.numeric(ncdf4::ncvar_get(measurement, "scan_acquisition_time", count = 1))
rndScan <- as.numeric(ncdf4::ncvar_get(measurement, "scan_acquisition_time", count = 10))[10]
rndScan2 <- as.numeric(ncdf4::ncvar_get(measurement, "scan_acquisition_time", count = 11))[11]
scans_per_second <- as.numeric((1/(rndScan2 - rndScan)))
if (isExact) {
full.matrix <- matrix(0, length(scan_indexes), ((max_mz - min_mz) + 1))
mass_values <- round(mass_values - (min_mz - 1))
for (i in 1:(length(scan_indexes) - 1)) {
MssLoc <- mass_values[(scan_indexes[i] + 1):scan_indexes[i + 1]]
MssInt <- mass_intensities[(scan_indexes[i] + 1):scan_indexes[i + 1]]
MssInt <- as.vector(unlist(lapply(split(MssInt, MssLoc), sum)))
MssLoc <- unique(MssLoc)
full.matrix[i, MssLoc] <- MssInt
}
}
else {
full.matrix <- matrix(0, length(scan_indexes), ((max_mz - min_mz) + 1))
mass_values <- mass_values - (min_mz - 1)
for (i in 1:(length(scan_indexes) - 1)) {
full.matrix[i, mass_values[(scan_indexes[i] + 1):scan_indexes[i + 1]]] <- mass_intensities[(scan_indexes[i] + 1):scan_indexes[i + 1]]
}
}
sampleRD <- new("RawDataParameters", data = full.matrix, min.mz = min_mz, max.mz = max_mz, start.time = start_time, mz.resolution = 1, scans.per.second = scans_per_second)
sampleRD
}
#' @name createdt
#' @aliases createdt
#' @title Creating Experiment Tables
#' @description eRah requires an instrumental and (optionally) phenotype .csv file for starting/creating a new eRah project/experiment. This function automatically creates the Phenoytpe and Instrumental data .csv files.
#' @usage createdt(path)
#' @param path the path where the experiment-folder is (where the experiment samples are stored).
#' @details
#' The experiment has to been organized as follows: all the samples related to each class have to be stored in the same folder (one folder = one class), and all the class-folders in one folder, which is the experiment folder.
#'
#' Two things have to be considered at this step: .csv files are different when created by American and European computers, so errors may raise due to that fact. Also, the folder containing the samples, must contain only folders. If the folder contains files (for example, already created .csv files), eRah will prompt an error.
#'
#' See eRah vignette for more details. To open the vignette, execute the following code in R:
#' vignette("eRahManual", package="erah")
#' @examples \dontrun{
#' # Store all the raw data files in one different folder per class,
#' # and all the class-folders in one folder, which is the experiment
#' # folder. Then execute
#'
#' createdt(path)
#'
#' # where path is the experiment folder path.
#' # The experiment can be now startd by:
#'
#' ex <- newExp(instrumental="path/DEMO_inst.csv",
#' phenotype="path/DEMO_pheno.csv", info="DEMO Experiment")
#' }
#' @seealso \code{\link{newExp}}
#' @export
#' @importFrom utils write.table
createdt <- function(path)
{
#path <- "Valli/GenCond"
#path.dir <- list.files(path)
path.name <- strsplit(path, "/")[[1]]
path.name <- path.name[length(path.name)]
#dirs.c <- unlist(apply(as.matrix(path.dir), 1, function(x) rep(x,length(list.files(paste(path, x, sep="/"))))))
#path.dir.c <- apply(as.matrix(path.dir),1, function(x) paste(path, x, sep="/"))
#files.c <- list.files(path.dir.c)
#files.name <- apply(as.matrix(1:length(dirs.c)),1, function(x) paste(dirs.c[x],files.c[x], sep="/"))
#files.name <- list.files(path.dir.c, full.name=T)
#files.ID <- apply(as.matrix(files.c), 1, function(x) strsplit(x, "\\.")[[1]][1])
files.name <- list.files(path, recursive=T)
if(any(apply(as.matrix(files.name), 1, function(x) length(strsplit(x, "/")[[1]]))==1)) stop("There are files without directory in the selected path. Remove all the files in the path, only folders are allowed")
files.class <- apply(as.matrix(files.name), 1, function(x) strsplit(x, "/")[[1]][1])
files.ID <- apply(as.matrix(files.name), 1, function(x) {
out.s <- strsplit(x, "/")[[1]]
strsplit(out.s[length(out.s)], "\\.")[[1]][1]
})
files.path <- apply(as.matrix(files.name),1, function(x) paste(path, x, sep="/"))
files.cdate <- apply(as.matrix(files.path), 1, function(x) as.character(file.info(x)$mtime))
files.date <- apply(as.matrix(files.cdate),1, function(x) strsplit(x, " ")[[1]][1])
files.time <- apply(as.matrix(files.cdate),1, function(x) strsplit(x, " ")[[1]][2])
inst.table <- matrix(0, ncol=4, nrow=length(files.ID))
colnames(inst.table) <- c("sampleID", "filename", "date", "time")
inst.table[,1] <- files.ID
inst.table[,2] <- files.name
inst.table[,3] <- files.date
inst.table[,4] <- files.time
meta.table <- matrix(0, ncol=2, nrow=length(files.ID))
colnames(meta.table) <- c("sampleID", "class")
meta.table[,1] <- files.ID
meta.table[,2] <- files.class
inst.file <- paste(path, "/", path.name, "_inst.csv", sep="")
meta.file <- paste(path, "/", path.name, "_pheno.csv", sep="")
write.table(inst.table, file=inst.file, sep=";", row.names=FALSE, eol="\n", quote=F)
write.table(meta.table, file=meta.file, sep=";", row.names=FALSE, eol="\n", quote=F)
}
#' @title Create Instrumental Table
#' @description Create table containing instrumental information such as sample IDs and file names.
#' @param files File paths to experiment samples.
#' @details Creates instrumental information table based on experiment sample file paths. Columns containing further information can also be added to this.
#' @examples \dontrun{
#' library(gcspikelite)
#'
#' files <- list.files(system.file('data',package = 'gcspikelite'),full.names = TRUE)
#' files <- files[sapply(files,grepl,pattern = 'CDF')]
#'
#' instrumental <- createInstrumentalTable(files)
#' }
#' @seealso \code{\link{newExp}} \code{\link{createPhenoTable}}
#' @importFrom tibble tibble
#' @export
createInstrumentalTable <- function(files){
files.ID <- apply(as.matrix(files), 1, function(x) {
out.s <- strsplit(x, "/")[[1]]
strsplit(out.s[length(out.s)], "\\.")[[1]][1]
})
files.cdate <- apply(as.matrix(files), 1, function(x) as.character(file.info(x)$mtime))
files.date <- apply(as.matrix(files.cdate),1, function(x) strsplit(x, " ")[[1]][1])
files.time <- apply(as.matrix(files.cdate),1, function(x) strsplit(x, " ")[[1]][2])
inst.table <- tibble(
sampleID = files.ID,
filename = files,
date = files.date,
time = files.time
)
return(inst.table)
}
#' @title Create Phenotype Table
#' @description Create table containing sample meta information such as as sample ID and class.
#' @param files File paths to experiment samples.
#' @param cls Character vector containing sample classes.
#' @details Creates phenotype information table based on experiment sample file paths and sample classes. Columns containing further information can also be added to this.
#' @examples \dontrun{
#' library(gcspikelite)
#' data(targets)
#'
#' files <- list.files(system.file('data',package = 'gcspikelite'),full.names = TRUE)
#' files <- files[sapply(files,grepl,pattern = 'CDF')]
#'
#' phenotype <- createPhenoTable(files,as.character(targets$Group[order(targets$FileName)]))
#' }
#' @seealso \code{\link{newExp}} \code{\link{createInstrumentalTable}}
#' @importFrom tibble tibble
#' @export
createPhenoTable <- function(files,cls){
files.ID <- apply(as.matrix(files), 1, function(x) {
out.s <- strsplit(x, "/")[[1]]
strsplit(out.s[length(out.s)], "\\.")[[1]][1]
})
meta.table <- tibble(
sampleID = files.ID,
class = cls
)
return(meta.table)
}