RPostClean.R
3.64 KB
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
#For Reading Raw Data from the created file
#Required Libraries
library(MASS)
library(dplyr)
library(tidyr)
library(readr)
library(stringr)
#Necessary Functions
#1# Function for discretizing the data
dndat <- function(NDATA){
rownd <- dim(NDATA)[1]
colnd <- dim(NDATA)[2]
DDATA <- matrix(0,nrow=rownd,ncol=colnd)
colnames(DDATA) <- colnames(NDATA)
i = 1
for(i in 1:rownd){
j <- 1
for(j in 1:colnd){
if(is.na(NDATA[i,j])==FALSE){
if(NDATA[i,j] < -1){
DDATA[i,j]=0L
}
if(NDATA[i,j] > 1){
DDATA[i,j]=2L
}
if(-1 <= NDATA[i,j] && NDATA[i,j] < 1){
DDATA[i,j]=1L
}
} else{
DDATA[i,j] = NDATA[i,j]
}
j = j + 1
}
i = i + 1
}
DDATA
}
#Bringing in the file
rawdat <- file.choose()
RAWDAT <- rawdat %>%
read_delim(delim ="\t",col_names = FALSE,skip=1) %>%
filter(.,!grepl("Group|Age|Region|PMI|Title|Sex|Braak",X1))
attributes(RAWDAT)$names <- RAWDAT[1,]
#Just the clinical data
RAWWORD <- rawdat %>%
read_delim(delim ="\t",col_names = FALSE,skip=1) %>%
filter(.,grepl("Group|Age|Region|PMI|Title|Sex|Braak",X1))
attributes(RAWWORD)$names <- RAWDAT[1,]
#Add col of NAs to clinical data
z <- 1
naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE)
for(z in 1:dim(RAWWORD)[1]){
naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,])))
z <- z + 1
}
colnames(naroww) <- "ROW_NAs"
RAWWORD <- bind_cols(RAWWORD,naroww)
##Getting back to the data
RAWDAT2 <- RAWDAT[-1,] %>%
dplyr::arrange(.,ID_REF)
##Editing the file for R processing
RAWDATID <- RAWDAT2[,1] %>%
as.matrix(.)
RAWDATNUM <- RAWDAT2[,-1] %>%
mapply(.,FUN = as.numeric) %>%
t(.)
##Consolidating genes with the same name
tabRDATID <- table(RAWDATID)
NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID))
j <- 1
for(j in 1:length(tabRDATID)){
##Putting the ones without duplicates in their new homes
if(tabRDATID[j] == 1){
NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])]
}
##Averaging duplicates and putting them in their new homes
if(tabRDATID[j] > 1){
NuRDATN[,j] <- rowMeans(RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])],na.rm = TRUE)
}
j <- j + 1
}
#Scaling the Data
scrawdat <- NuRDATN%>%
scale()
attr(scrawdat,"scaled:center") <- NULL
attr(scrawdat,"scaled:scale") <- NULL
colnames(scrawdat) <- rownames(tabRDATID)
#Discretized the Data
dialzdat <- scrawdat %>%
dndat(.) %>%
t()%>%
as.data.frame(.)
colnames(dialzdat) <- rownames(RAWDATNUM)
#gene names
genena <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1))
#setting "ID_REF" as a new variable
colnames(genena) <- "ID_REF"
rownames(dialzdat) <- NULL
dialzdat <-bind_cols(genena,dialzdat)
#NAs in a column
x <- 2
nacol <- as.data.frame(t(rep(0,dim(dialzdat)[2])),stringsAsFactors = FALSE)
nacol[1,1] = "COL_NAs"
for(x in 2:dim(dialzdat)[2]){
nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x])))
x <- x + 1
}
colnames(nacol) <- colnames(dialzdat)
dialzdat<-bind_rows(dialzdat,nacol)
#NAs in a row
y <- 1
narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE)
for(y in 1:dim(dialzdat)[1]){
narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,])))
y <- y + 1
}
colnames(narowd) <- "ROW_NAs"
dialzdat <- bind_cols(dialzdat,narowd)
#converting to character so that the clinical can be brought together with discrete data
k <- 2
for(k in 2:dim(dialzdat)[2]-1){
dialzdat[,k] <- as.character(dialzdat[,k])
k <- k + 1
}
#The End the full data
Fullalzdw <- bind_rows(RAWWORD,dialzdat)
#Create the file
nfnaex <- strsplit(rawdat,"[\\|/]") %>%
.[[1]] %>%
.[length(.)] %>%
gsub("\\D","",.) %>%
c("GSE",.,"dscrt.txt") %>%
paste(collapse = "")
write.table(Fullalzdw, file = nfnaex, sep = "\t",col.names = TRUE,row.names = FALSE)