Efrain Gonzalez / Cleaning and Fixing Data with R

Download as
Browse Code ยป

Commit 769f081f9b31cbc0a244813efb135609926f88cf

Authored by Efrain Gonzalez
1 parent c9b2687075
Exists in master

Don't use this code yet 

update to discretization part
Showing 1 changed file with 2 additions and 2 deletions   Show diff stats
1 ######################################################################## 1 ########################################################################
2 # Don't Use This Code Just Yet # 2 # Don't Use This Code Just Yet #
3 ######################################################################## 3 ########################################################################
4 #Efrain H. Gonzalez 4 #Efrain H. Gonzalez
5 #6/16/2017 5 #6/16/2017
6 6
7 #Libraries required to run the code 7 #Libraries required to run the code
8 library(pryr) 8 library(pryr)
9 library(MASS) 9 library(MASS)
10 library(dplyr) 10 library(dplyr)
11 library(tidyr) 11 library(tidyr)
12 library(readr) 12 library(readr)
13 library(stringr) 13 library(stringr)
14 14
15 15
16 #Necessary Functions 16 #Necessary Functions
17 #1#Function for handling the changing of row names and column names 17 #1#Function for handling the changing of row names and column names
18 chngrownm <- function(mat){ 18 chngrownm <- function(mat){
19 row <- dim(mat)[1] 19 row <- dim(mat)[1]
20 col <- dim(mat)[2] 20 col <- dim(mat)[2]
21 e <- 1 21 e <- 1
22 r <- 1 22 r <- 1
23 a <- 1 23 a <- 1
24 h <- 1 24 h <- 1
25 g <- 1 25 g <- 1
26 o <- 1 26 o <- 1
27 for(e in 1:col){ 27 for(e in 1:col){
28 if("!Sample_source_name_ch1"==mat[1,e]){ 28 if("!Sample_source_name_ch1"==mat[1,e]){
29 colnames(mat)[e] <- "Brain_Region" 29 colnames(mat)[e] <- "Brain_Region"
30 } 30 }
31 else if("!Sample_title" == mat[1,e]){ 31 else if("!Sample_title" == mat[1,e]){
32 colnames(mat)[e] <- "Title" 32 colnames(mat)[e] <- "Title"
33 } 33 }
34 else if("!Sample_geo_accession" == mat[1,e]){ 34 else if("!Sample_geo_accession" == mat[1,e]){
35 colnames(mat)[e] <- "ID_REF" 35 colnames(mat)[e] <- "ID_REF"
36 } else{ 36 } else{
37 if(grepl("Sex|gender|Gender|sex",mat[2,e])==TRUE){ 37 if(grepl("Sex|gender|Gender|sex",mat[2,e])==TRUE){
38 colnames(mat)[e] <- paste0("Sex",r) 38 colnames(mat)[e] <- paste0("Sex",r)
39 r = r + 1 39 r = r + 1
40 } 40 }
41 if(grepl("postmorteminterval|PMI|pmi|interval",mat[2,e])==TRUE){ 41 if(grepl("postmorteminterval|PMI|pmi|interval",mat[2,e])==TRUE){
42 colnames(mat)[e] <- paste0("PMI",a) 42 colnames(mat)[e] <- paste0("PMI",a)
43 a = a + 1 43 a = a + 1
44 } 44 }
45 if(grepl("age|Age|AGE",mat[2,e])==TRUE){ 45 if(grepl("age|Age|AGE",mat[2,e])==TRUE){
46 colnames(mat)[e] <- paste0("Age",h) 46 colnames(mat)[e] <- paste0("Age",h)
47 h = h + 1 47 h = h + 1
48 } 48 }
49 if(grepl("braak|b&b",mat[2,e])==TRUE){ 49 if(grepl("braak|b&b",mat[2,e])==TRUE){
50 colnames(mat)[e] <- paste0("Braak",g) 50 colnames(mat)[e] <- paste0("Braak",g)
51 g = g + 1 51 g = g + 1
52 } 52 }
53 if(grepl("group|disease|control|AD|normal|diagnosis|Alzheimer|Control|Normal",mat[2,e])==TRUE){ 53 if(grepl("group|disease|control|AD|normal|diagnosis|Alzheimer|Control|Normal",mat[2,e])==TRUE){
54 colnames(mat)[e] <- paste0("Group",o) 54 colnames(mat)[e] <- paste0("Group",o)
55 o = o + 1 55 o = o + 1
56 } 56 }
57 57
58 } 58 }
59 e = e + 1 59 e = e + 1
60 } 60 }
61 mat 61 mat
62 } 62 }
63 63
64 #2#Function for reorganizing information within the columns 64 #2#Function for reorganizing information within the columns
65 cinfo <- function(mat){ 65 cinfo <- function(mat){
66 col <- dim(mat)[2] 66 col <- dim(mat)[2]
67 j <-2 67 j <-2
68 for(j in 2:col){ 68 for(j in 2:col){
69 if(grepl("Group",colnames(mat)[j]) == TRUE){ 69 if(grepl("Group",colnames(mat)[j]) == TRUE){
70 mat[,j] <- gsub(".+:\\s|\\s.+;.+","",mat[,j]) 70 mat[,j] <- gsub(".+:\\s|\\s.+;.+","",mat[,j])
71 } 71 }
72 else if(grepl("Age",colnames(mat)[j])==TRUE){ 72 else if(grepl("Age",colnames(mat)[j])==TRUE){
73 mat[,j] <- gsub("\\D","",mat[,j])%>% 73 mat[,j] <- gsub("\\D","",mat[,j])%>%
74 as.integer() 74 as.integer()
75 } 75 }
76 else if(grepl("Sex",colnames(mat)[j])==TRUE){ 76 else if(grepl("Sex",colnames(mat)[j])==TRUE){
77 mat[,j] <- gsub(".+:\\s","",mat[,j]) 77 mat[,j] <- gsub(".+:\\s","",mat[,j])
78 } 78 }
79 else if(grepl("PMI",colnames(mat)[j])==TRUE){ 79 else if(grepl("PMI",colnames(mat)[j])==TRUE){
80 mat[,j] <- gsub("[^0-9\\.]","",mat[,j])%>% 80 mat[,j] <- gsub("[^0-9\\.]","",mat[,j])%>%
81 as.numeric() 81 as.numeric()
82 } 82 }
83 else if(grepl("Braak",colnames(mat)[j])==TRUE){ 83 else if(grepl("Braak",colnames(mat)[j])==TRUE){
84 mat[,j]<-gsub(".+:\\s","",mat[,j])%>% 84 mat[,j]<-gsub(".+:\\s","",mat[,j])%>%
85 as.roman()%>% 85 as.roman()%>%
86 as.integer() 86 as.integer()
87 } 87 }
88 j=j+1 88 j=j+1
89 } 89 }
90 mat 90 mat
91 } 91 }
92 92
93 #3#Function for labeling the gene IDs without names 93 #3#Function for labeling the gene IDs without names
94 NAFIXING <- function(GIDNAM){ 94 NAFIXING <- function(GIDNAM){
95 row <- dim(GIDNAM)[1] 95 row <- dim(GIDNAM)[1]
96 i <- 1 96 i <- 1
97 for(i in 1:row){ 97 for(i in 1:row){
98 if(grepl("^NA\\s*$",GIDNAM[i,2])==TRUE||is.na(GIDNAM[i,2])==TRUE){ 98 if(grepl("^NA\\s*$",GIDNAM[i,2])==TRUE||is.na(GIDNAM[i,2])==TRUE){
99 GIDNAM[i,2] <- GIDNAM[i,1] 99 GIDNAM[i,2] <- GIDNAM[i,1]
100 } 100 }
101 i <- i + 1 101 i <- i + 1
102 } 102 }
103 GIDNAM 103 GIDNAM
104 } 104 }
105 105
106 #4#Function for changing the gene ID to gene name 106 #4#Function for changing the gene ID to gene name
107 cgeneID <- function(GeneName,DATA){ 107 cgeneID <- function(GeneName,DATA){
108 nj <- t(GeneName) 108 nj <- t(GeneName)
109 nq <- t(DATA) 109 nq <- t(DATA)
110 colGene <- dim(nj)[2] 110 colGene <- dim(nj)[2]
111 colDATA <- dim(nq)[2] 111 colDATA <- dim(nq)[2]
112 j <- 1 112 j <- 1
113 for(j in 1:colDATA){ 113 for(j in 1:colDATA){
114 #where is that gene id located within the GPL file 114 #where is that gene id located within the GPL file
115 chngreq <- grep(paste0("^",nq[1,j],"$"),nj[1,]) 115 chngreq <- grep(paste0("^",nq[1,j],"$"),nj[1,])
116 if(is.na(sum(chngreq))==FALSE){ 116 if(is.na(sum(chngreq))==FALSE){
117 if(sum(chngreq) > 0){ 117 if(sum(chngreq) > 0){
118 nq[1,j] <- gsub(paste0("^",nq[1,j],"$"),nj[2,chngreq],nq[1,j]) 118 nq[1,j] <- gsub(paste0("^",nq[1,j],"$"),nj[2,chngreq],nq[1,j])
119 } 119 }
120 } 120 }
121 j <- j + 1 121 j <- j + 1
122 } 122 }
123 nq 123 nq
124 } 124 }
125 #cgeneID <- function(GeneName,DATA){ 125 #cgeneID <- function(GeneName,DATA){
126 # colGene <- dim(GeneName)[2] 126 # colGene <- dim(GeneName)[2]
127 # j <- 1 127 # j <- 1
128 # for(j in 1:colGene){ 128 # for(j in 1:colGene){
129 # chngsreq <- grep(paste0("^",GeneName[1,j],"$"),DATA[1,]) 129 # chngsreq <- grep(paste0("^",GeneName[1,j],"$"),DATA[1,])
130 # if(is.na(sum(chngsreq))==FALSE){ 130 # if(is.na(sum(chngsreq))==FALSE){
131 # if(sum(chngsreq) > 0){ 131 # if(sum(chngsreq) > 0){
132 # DATA[1,chngsreq] <- gsub(paste0("^",GeneName[1,j]),GeneName[2,j],DATA[1,chngsreq]) 132 # DATA[1,chngsreq] <- gsub(paste0("^",GeneName[1,j]),GeneName[2,j],DATA[1,chngsreq])
133 # } 133 # }
134 # } 134 # }
135 # j = j+1 135 # j = j+1
136 # } 136 # }
137 # DATA 137 # DATA
138 #} 138 #}
139 139
140 #5#Function for adjusting the gene names 140 #5#Function for adjusting the gene names
141 gcnames <- function(DiData,usecol=1){ 141 gcnames <- function(DiData,usecol=1){
142 nuruns <- dim(DiData)[2] 142 nuruns <- dim(DiData)[2]
143 i = 1 143 i = 1
144 nwnam <- rep("0",length.out=nuruns) 144 nwnam <- rep("0",length.out=nuruns)
145 for(i in 1:nuruns){ 145 for(i in 1:nuruns){
146 if(length(strsplit(colnames(DiData)[i],"///")[[1]]) >= usecol){ 146 if(length(strsplit(colnames(DiData)[i],"///")[[1]]) >= usecol){
147 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][usecol]) 147 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][usecol])
148 } else{ 148 } else{
149 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][1]) 149 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][1])
150 } 150 }
151 151
152 } 152 }
153 nwnam 153 nwnam
154 154
155 } 155 }
156 156
157 #6# Function for discretizing the data 157 #6# Function for discretizing the data
158 dndat <- function(NDATA){ 158 dndat <- function(NDATA){
159 rownd <- dim(NDATA)[1] 159 rownd <- dim(NDATA)[1]
160 colnd <- dim(NDATA)[2] 160 colnd <- dim(NDATA)[2]
161 DDATA <- matrix(0,nrow=rownd,ncol=colnd) 161 DDATA <- matrix(0,nrow=rownd,ncol=colnd)
162 colnames(DDATA) <- colnames(NDATA) 162 colnames(DDATA) <- colnames(NDATA)
163 i <- 1 163 i <- 1
164 for(i in 1:rownd){ 164 for(i in 1:rownd){
165 j <- 1 165 j <- 1
166 for(j in 1:colnd){ 166 for(j in 1:colnd){
167 if(is.na(NDATA[i,j])==FALSE){ 167 if(is.na(NDATA[i,j])==FALSE){
168 168
169 if(NDATA[i,j] < -1){ 169 if(NDATA[i,j] < -1){
170 DDATA[i,j]=0L 170 DDATA[i,j]=0L
171 } 171 }
172 if(NDATA[i,j] > 1){ 172 else if(NDATA[i,j] > 1){
173 DDATA[i,j]=2L 173 DDATA[i,j]=2L
174 } 174 }
175 if(-1 <= NDATA[i,j] && NDATA[i,j] < 1){ 175 else if(-1 <= NDATA[i,j] && NDATA[i,j] < 1){
176 DDATA[i,j]=1L 176 DDATA[i,j]=1L
177 } 177 }
178 } else{ 178 } else{
179 DDATA[i,j] = NDATA[i,j] 179 DDATA[i,j] = NDATA[i,j]
180 } 180 }
181 j = j + 1 181 j = j + 1
182 } 182 }
183 i = i + 1 183 i = i + 1
184 } 184 }
185 DDATA 185 DDATA
186 } 186 }
187 187
188 188
189 #MajorFunction#This is the function that does everything else 189 #MajorFunction#This is the function that does everything else
190 THEFT <- function(){ 190 THEFT <- function(){
191 #Set working directory based on the directory of the series matrix file Currently only works for windows 191 #Set working directory based on the directory of the series matrix file Currently only works for windows
192 wd <- getwd() 192 wd <- getwd()
193 #list.files() 193 #list.files()
194 #gsub("wd",wd,"Do you want to clean all data files in the directory wd?") 194 #gsub("wd",wd,"Do you want to clean all data files in the directory wd?")
195 numDAT <- switch(EXPR = menu(choices = c("Yes","No"),title = gsub("wd",wd,"Do you want to clean all data files in the directory wd?")) + 1,cat("Nothing done\n"),1L,2L) 195 numDAT <- switch(EXPR = menu(choices = c("Yes","No"),title = gsub("wd",wd,"Do you want to clean all data files in the directory wd?")) + 1,cat("Nothing done\n"),1L,2L)
196 GSEfileloc <- grep("^GSE.+\\.txt\\.gz$",list.files()) 196 GSEfileloc <- grep("^GSE.+\\.txt\\.gz$",list.files())
197 GSEfloc <- list.files()[GSEfileloc] 197 GSEfloc <- list.files()[GSEfileloc]
198 #ALL DATA FILES WILL BE CLEANED 198 #ALL DATA FILES WILL BE CLEANED
199 if(numDAT == 1){ 199 if(numDAT == 1){
200 #indexing the data files 200 #indexing the data files
201 n <- 1 201 n <- 1
202 for(n in 1: length(GSEfloc)){ 202 for(n in 1: length(GSEfloc)){
203 alz <- GSEfloc[n] 203 alz <- GSEfloc[n]
204 204
205 #Working with the wordy part of the document 205 #Working with the wordy part of the document
206 alzword <- alz %>% 206 alzword <- alz %>%
207 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>% 207 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>%
208 filter(grepl("!Sample",X1))%>% 208 filter(grepl("!Sample",X1))%>%
209 filter(!grepl("!Sample_contact",X1)) 209 filter(!grepl("!Sample_contact",X1))
210 210
211 #Getting the GPL file 211 #Getting the GPL file
212 genena <- grep("_platform_id",alzword$X1) %>% 212 genena <- grep("_platform_id",alzword$X1) %>%
213 alzword$X2[.] %>% 213 alzword$X2[.] %>%
214 str_trim(.) %>% 214 str_trim(.) %>%
215 paste0("^",.,"\\D") %>% 215 paste0("^",.,"\\D") %>%
216 grep(.,list.files()) %>% 216 grep(.,list.files()) %>%
217 list.files()[.] 217 list.files()[.]
218 218
219 #Find out if it is a soft GPL file or not 219 #Find out if it is a soft GPL file or not
220 soft <- strsplit(genena,"[\\|/]") %>% 220 soft <- strsplit(genena,"[\\|/]") %>%
221 .[[1]] %>% 221 .[[1]] %>%
222 .[length(.)] %>% 222 .[length(.)] %>%
223 grepl("soft",.) 223 grepl("soft",.)
224 224
225 ##Changing row names and column names: 225 ##Changing row names and column names:
226 ALZWORD <- t(alzword) 226 ALZWORD <- t(alzword)
227 rownames(ALZWORD)=NULL 227 rownames(ALZWORD)=NULL
228 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE) 228 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE)
229 ALZWORD <- chngrownm(ALZWORD)[-1,] 229 ALZWORD <- chngrownm(ALZWORD)[-1,]
230 ALZWORD <- ALZWORD%>% 230 ALZWORD <- ALZWORD%>%
231 as.data.frame()%>% 231 as.data.frame()%>%
232 dplyr::select(-starts_with("col")) 232 dplyr::select(-starts_with("col"))
233 233
234 ##Reorganizing information within the columns and final clinical data 234 ##Reorganizing information within the columns and final clinical data
235 ALZWORDF <- cinfo(ALZWORD) 235 ALZWORDF <- cinfo(ALZWORD)
236 236
237 237
238 #Working with Actual Data part of file 238 #Working with Actual Data part of file
239 alzdat <- alz %>% 239 alzdat <- alz %>%
240 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1) 240 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1)
241 ALZDAT <- t(alzdat[,-1]) 241 ALZDAT <- t(alzdat[,-1])
242 rownames(ALZDAT)=NULL 242 rownames(ALZDAT)=NULL
243 243
244 ##Is there a clean version of the GPL file available? 244 ##Is there a clean version of the GPL file available?
245 gplnum <- strsplit(genena,"[\\|/]") %>% 245 gplnum <- strsplit(genena,"[\\|/]") %>%
246 .[[1]] %>% 246 .[[1]] %>%
247 .[length(.)] %>% 247 .[length(.)] %>%
248 gsub("\\D","",.) 248 gsub("\\D","",.)
249 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files())) 249 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files()))
250 if(clfileex >= 1){ 250 if(clfileex >= 1){
251 #use the clean version 251 #use the clean version
252 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>% 252 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>%
253 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!") 253 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!")
254 254
255 } 255 }
256 else if(clfileex == 0){ 256 else if(clfileex == 0){
257 ##Lets Create a clean version 257 ##Lets Create a clean version
258 258
259 ##Gene ID to Gene Name 259 ##Gene ID to Gene Name
260 if(soft == TRUE){ 260 if(soft == TRUE){
261 #Check to see if there is already a file containing information on soft files 261 #Check to see if there is already a file containing information on soft files
262 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files())) 262 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files()))
263 if(fileex == 1){ 263 if(fileex == 1){
264 #Check to see if this GPL soft file has been used before 264 #Check to see if this GPL soft file has been used before
265 IDF <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 265 IDF <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
266 .$GPL_FILE_NUM%>% 266 .$GPL_FILE_NUM%>%
267 grepl(gplnum,.) %>% 267 grepl(gplnum,.) %>%
268 sum() 268 sum()
269 if(IDF == 1){ 269 if(IDF == 1){
270 IDLOCAL <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 270 IDLOCAL <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
271 .$GPL_FILE_NUM%>% 271 .$GPL_FILE_NUM%>%
272 grep(gplnum,.) 272 grep(gplnum,.)
273 idlocgpl <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 273 idlocgpl <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
274 .$LOC_ID %>% 274 .$LOC_ID %>%
275 .[IDLOCAL] 275 .[IDLOCAL]
276 geneIDNam <- genena %>% 276 geneIDNam <- genena %>%
277 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>% 277 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>%
278 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 278 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
279 } 279 }
280 else if(IDF == 0){ 280 else if(IDF == 0){
281 #No information on this particular GPL file 281 #No information on this particular GPL file
282 idLOCGPL <- genena %>% 282 idLOCGPL <- genena %>%
283 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 283 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
284 t(.) %>% 284 t(.) %>%
285 grep("^ID\\s*$",.) %>% 285 grep("^ID\\s*$",.) %>%
286 -1 286 -1
287 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>% 287 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>%
288 cat(file="GPL_ID_LOC.txt",sep = "\t", fill = TRUE, append = TRUE) 288 cat(file="GPL_ID_LOC.txt",sep = "\t", fill = TRUE, append = TRUE)
289 geneIDNam <- genena %>% 289 geneIDNam <- genena %>%
290 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 290 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
291 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 291 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
292 } 292 }
293 } 293 }
294 else if(fileex == 0){ 294 else if(fileex == 0){
295 #We must create a file that we can access for later use 295 #We must create a file that we can access for later use
296 idLOCGPL <- genena %>% 296 idLOCGPL <- genena %>%
297 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 297 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
298 t(.) %>% 298 t(.) %>%
299 grep("^ID\\s*$",.) %>% 299 grep("^ID\\s*$",.) %>%
300 -1 300 -1
301 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL)) 301 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL))
302 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID") 302 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID")
303 write.table(Firstval,file = "GPL_ID_LOC.txt", sep = "\t",row.names = FALSE, col.names = TRUE) 303 write.table(Firstval,file = "GPL_ID_LOC.txt", sep = "\t",row.names = FALSE, col.names = TRUE)
304 geneIDNam <- genena %>% 304 geneIDNam <- genena %>%
305 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 305 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
306 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 306 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
307 } 307 }
308 } 308 }
309 else if(soft == FALSE){ 309 else if(soft == FALSE){
310 geneIDNam <- genena %>% 310 geneIDNam <- genena %>%
311 read_delim(delim="\t",comment = "#")%>% 311 read_delim(delim="\t",comment = "#")%>%
312 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 312 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
313 } 313 }
314 314
315 ##Labeling the gene IDs without names 315 ##Labeling the gene IDs without names
316 geneIDNam <- NAFIXING(geneIDNam) 316 geneIDNam <- NAFIXING(geneIDNam)
317 317
318 ##remove the whitespace 318 ##remove the whitespace
319 geneIDNam <- t(rbind(str_trim(t(geneIDNam)[1,]),str_trim(t(geneIDNam)[2,]))) 319 geneIDNam <- t(rbind(str_trim(t(geneIDNam)[1,]),str_trim(t(geneIDNam)[2,])))
320 320
321 ##Here is the clean version 321 ##Here is the clean version
322 write.table(geneIDNam,file = paste0("Clean_GPL",gplnum,".txt"),sep = "\t",row.names = FALSE, col.names = FALSE) 322 write.table(geneIDNam,file = paste0("Clean_GPL",gplnum,".txt"),sep = "\t",row.names = FALSE, col.names = FALSE)
323 } 323 }
324 324
325 325
326 326
327 ##Changing the gene ID to gene name 327 ##Changing the gene ID to gene name
328 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat)) 328 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat))
329 colnames(ALZDAT) = ALZDAT1[1,] 329 colnames(ALZDAT) = ALZDAT1[1,]
330 330
331 331
332 ##Adjusting the column names aka the gene names 332 ##Adjusting the column names aka the gene names
333 colnames(ALZDAT) <- gcnames(ALZDAT) 333 colnames(ALZDAT) <- gcnames(ALZDAT)
334 334
335 335
336 #Full RAW Data 336 #Full RAW Data
337 Fullalzdwr <- ALZDAT %>% 337 Fullalzdwr <- ALZDAT %>%
338 as.data.frame() %>% 338 as.data.frame() %>%
339 cbind(ALZWORDF,.) 339 cbind(ALZWORDF,.)
340 340
341 #Raw file is output 341 #Raw file is output
342 nfnaex <- strsplit(alz,"[\\]") %>% 342 nfnaex <- strsplit(alz,"[\\]") %>%
343 .[[1]] %>% 343 .[[1]] %>%
344 .[length(.)] %>% 344 .[length(.)] %>%
345 gsub("\\D","",.) %>% 345 gsub("\\D","",.) %>%
346 c("GSE",.,"aftexcel.txt") %>% 346 c("GSE",.,"aftexcel.txt") %>%
347 paste(collapse = "") 347 paste(collapse = "")
348 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t") 348 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t")
349 349
350 350
351 351
352 #Now for the discretization part 352 #Now for the discretization part
353 ##get the wordy part again 353 ##get the wordy part again
354 rawword <- t(ALZWORDF) 354 rawword <- t(ALZWORDF)
355 355
356 ##where is ID_REF located 356 ##where is ID_REF located
357 hereim <- grep("ID_REF",rownames(rawword)) 357 hereim <- grep("ID_REF",rownames(rawword))
358 358
359 ##Subject Names GSM... 359 ##Subject Names GSM...
360 subjnam <- rawword[hereim,] 360 subjnam <- rawword[hereim,]
361 361
362 ##Getting the names for the rows 362 ##Getting the names for the rows
363 namedarows <- rownames(rawword)[-hereim] %>% 363 namedarows <- rownames(rawword)[-hereim] %>%
364 as.data.frame() 364 as.data.frame()
365 RAWWORD <- rawword[-hereim,] %>% 365 RAWWORD <- rawword[-hereim,] %>%
366 as.data.frame() %>% 366 as.data.frame() %>%
367 bind_cols(namedarows,.) 367 bind_cols(namedarows,.)
368 z <- 1 368 z <- 1
369 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE) 369 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE)
370 for(z in 1:dim(RAWWORD)[1]){ 370 for(z in 1:dim(RAWWORD)[1]){
371 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,]))) 371 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,])))
372 z <- z + 1 372 z <- z + 1
373 } 373 }
374 374
375 colnames(naroww) <- "ROW_NAs" 375 colnames(naroww) <- "ROW_NAs"
376 RAWWORD <- bind_cols(RAWWORD,naroww) 376 RAWWORD <- bind_cols(RAWWORD,naroww)
377 377
378 378
379 roALZna <- t(ALZDAT) %>% 379 roALZna <- t(ALZDAT) %>%
380 rownames(.) %>% 380 rownames(.) %>%
381 as.data.frame(.) 381 as.data.frame(.)
382 colnames(roALZna) <- "ID_REF" 382 colnames(roALZna) <- "ID_REF"
383 383
384 RAWDAT <- t(ALZDAT) %>% 384 RAWDAT <- t(ALZDAT) %>%
385 as.data.frame(.) 385 as.data.frame(.)
386 colnames(RAWDAT) <- NULL 386 colnames(RAWDAT) <- NULL
387 rownames(RAWDAT) <- NULL 387 rownames(RAWDAT) <- NULL
388 388
389 RAWDAT2 <- RAWDAT %>% 389 RAWDAT2 <- RAWDAT %>%
390 cbind(roALZna,.) %>% 390 cbind(roALZna,.) %>%
391 dplyr::arrange(.,ID_REF) 391 dplyr::arrange(.,ID_REF)
392 392
393 ##Editing the file for R processing 393 ##Editing the file for R processing
394 RAWDATID <- RAWDAT2[,1] %>% 394 RAWDATID <- RAWDAT2[,1] %>%
395 as.matrix(.) 395 as.matrix(.)
396 396
397 RAWDATNUM <- RAWDAT2[,-1] %>% 397 RAWDATNUM <- RAWDAT2[,-1] %>%
398 mapply(.,FUN = as.numeric) %>% 398 mapply(.,FUN = as.numeric) %>%
399 t(.) 399 t(.)
400 400
401 ##Consolidating genes with the same name 401 ##Consolidating genes with the same name
402 ###create empty matrix of size equal to tabRDATID 402 ###create empty matrix of size equal to tabRDATID
403 tabRDATID <- table(RAWDATID) 403 tabRDATID <- table(RAWDATID)
404 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID)) 404 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID))
405 j <- 1 405 j <- 1
406 for(j in 1:length(tabRDATID)){ 406 for(j in 1:length(tabRDATID)){
407 ##Putting the ones without duplicates in their new homes 407 ##Putting the ones without duplicates in their new homes
408 if(tabRDATID[j] == 1){ 408 if(tabRDATID[j] == 1){
409 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])] 409 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])]
410 } 410 }
411 ##Averaging duplicates and putting them in their new homes 411 ##Averaging duplicates and putting them in their new homes
412 else if(tabRDATID[j] > 1){ 412 else if(tabRDATID[j] > 1){
413 NuRDATN[,j] <- rowMeans(RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])],na.rm = TRUE) 413 NuRDATN[,j] <- rowMeans(RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])],na.rm = TRUE)
414 } 414 }
415 j <- j + 1 415 j <- j + 1
416 } 416 }
417 417
418 ##Scaling the Data 418 ##Scaling the Data
419 scrawdat <- NuRDATN%>% 419 scrawdat <- NuRDATN%>%
420 scale() 420 scale()
421 attr(scrawdat,"scaled:center") <- NULL 421 attr(scrawdat,"scaled:center") <- NULL
422 attr(scrawdat,"scaled:scale") <- NULL 422 attr(scrawdat,"scaled:scale") <- NULL
423 colnames(scrawdat) <- rownames(tabRDATID) 423 colnames(scrawdat) <- rownames(tabRDATID)
424 424
425 ##Discretized the Data 425 ##Discretized the Data
426 dialzdat <- scrawdat %>% 426 dialzdat <- scrawdat %>%
427 dndat(.) %>% 427 dndat(.) %>%
428 t()%>% 428 t()%>%
429 as.data.frame(.) 429 as.data.frame(.)
430 colnames(dialzdat) <- rownames(RAWDATNUM) 430 colnames(dialzdat) <- rownames(RAWDATNUM)
431 431
432 ##setting "ID_REF" as a new variable 432 ##setting "ID_REF" as a new variable
433 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1)) 433 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1))
434 colnames(geneNAM) <- "ID_REF" 434 colnames(geneNAM) <- "ID_REF"
435 rownames(dialzdat) <- NULL 435 rownames(dialzdat) <- NULL
436 dialzdat <-bind_cols(geneNAM,dialzdat) 436 dialzdat <-bind_cols(geneNAM,dialzdat)
437 437
438 ##NAs in a column 438 ##NAs in a column
439 x <- 2 439 x <- 2
440 nacol <- as.data.frame(t(rep(0,dim(dialzdat)[2])),stringsAsFactors = FALSE) 440 nacol <- as.data.frame(t(rep(0,dim(dialzdat)[2])),stringsAsFactors = FALSE)
441 nacol[1,1] = "COL_NAs" 441 nacol[1,1] = "COL_NAs"
442 for(x in 2:dim(dialzdat)[2]){ 442 for(x in 2:dim(dialzdat)[2]){
443 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x]))) 443 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x])))
444 x <- x + 1 444 x <- x + 1
445 } 445 }
446 colnames(nacol) <- colnames(dialzdat) 446 colnames(nacol) <- colnames(dialzdat)
447 dialzdat <- bind_rows(dialzdat,nacol) 447 dialzdat <- bind_rows(dialzdat,nacol)
448 448
449 ##NAs in a row 449 ##NAs in a row
450 y <- 1 450 y <- 1
451 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE) 451 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE)
452 for(y in 1:dim(dialzdat)[1]){ 452 for(y in 1:dim(dialzdat)[1]){
453 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,]))) 453 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,])))
454 y <- y + 1 454 y <- y + 1
455 } 455 }
456 colnames(narowd) <- "ROW_NAs" 456 colnames(narowd) <- "ROW_NAs"
457 dialzdat <- bind_cols(dialzdat,narowd) 457 dialzdat <- bind_cols(dialzdat,narowd)
458 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam 458 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam
459 colnames(RAWWORD) <- colnames(dialzdat) 459 colnames(RAWWORD) <- colnames(dialzdat)
460 ##converting to character so that the clinical can be brought together with discrete data 460 ##converting to character so that the clinical can be brought together with discrete data
461 k <- 2 461 k <- 2
462 for(k in 2:dim(dialzdat)[2]-1){ 462 for(k in 2:dim(dialzdat)[2]-1){
463 dialzdat[,k] <- as.character(dialzdat[,k]) 463 dialzdat[,k] <- as.character(dialzdat[,k])
464 k <- k + 1 464 k <- k + 1
465 } 465 }
466 #The End the full data 466 #The End the full data
467 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat) 467 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat)
468 468
469 #Produces Discrete file 469 #Produces Discrete file
470 nfnaex2 <- strsplit(alz,"[\\|/]") %>% 470 nfnaex2 <- strsplit(alz,"[\\|/]") %>%
471 .[[1]] %>% 471 .[[1]] %>%
472 .[length(.)] %>% 472 .[length(.)] %>%
473 gsub("\\D","",.) %>% 473 gsub("\\D","",.) %>%
474 c("GSE",.,"dscrt.txt") %>% 474 c("GSE",.,"dscrt.txt") %>%
475 paste(collapse = "") 475 paste(collapse = "")
476 write.table(Dscrtalzdw, file = nfnaex2, sep = "\t",col.names = TRUE,row.names = FALSE) 476 write.table(Dscrtalzdw, file = nfnaex2, sep = "\t",col.names = TRUE,row.names = FALSE)
477 n <- n +1 477 n <- n +1
478 } 478 }
479 } 479 }
480 480
481 #CHOOSE A DATA FILE TO CLEAN OR SEVERAL DATA FILES TO CLEAN 481 #CHOOSE A DATA FILE TO CLEAN OR SEVERAL DATA FILES TO CLEAN
482 else if(numDAT == 2){ 482 else if(numDAT == 2){
483 #All the files you want to analyze 483 #All the files you want to analyze
484 ANDIS <- select.list(choices = list.files()[GSEfileloc],multiple = TRUE, title = "Choose the file/files you want to analyze:") 484 ANDIS <- select.list(choices = list.files()[GSEfileloc],multiple = TRUE, title = "Choose the file/files you want to analyze:")
485 if(length(ANDIS) == 0){ 485 if(length(ANDIS) == 0){
486 #Spit out a warning 486 #Spit out a warning
487 warning("You did not select any files and so no cleaning will be performed") 487 warning("You did not select any files and so no cleaning will be performed")
488 } else{ 488 } else{
489 #indexing the data files 489 #indexing the data files
490 n <- 1 490 n <- 1
491 for(n in 1: length(ANDIS)){ 491 for(n in 1: length(ANDIS)){
492 alz <- ANDIS[n] 492 alz <- ANDIS[n]
493 493
494 #Working with the wordy part of the document 494 #Working with the wordy part of the document
495 alzword <- alz %>% 495 alzword <- alz %>%
496 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>% 496 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>%
497 filter(grepl("!Sample",X1))%>% 497 filter(grepl("!Sample",X1))%>%
498 filter(!grepl("!Sample_contact",X1)) 498 filter(!grepl("!Sample_contact",X1))
499 499
500 #Getting the GPL file 500 #Getting the GPL file
501 genena <- grep("_platform_id",alzword$X1) %>% 501 genena <- grep("_platform_id",alzword$X1) %>%
502 alzword$X2[.] %>% 502 alzword$X2[.] %>%
503 str_trim(.) %>% 503 str_trim(.) %>%
504 paste0("^",.,"\\D") %>% 504 paste0("^",.,"\\D") %>%
505 grep(.,list.files()) %>% 505 grep(.,list.files()) %>%
506 list.files()[.] 506 list.files()[.]
507 507
508 #Find out if it is a soft GPL file or not 508 #Find out if it is a soft GPL file or not
509 soft <- strsplit(genena,"[\\|/]") %>% 509 soft <- strsplit(genena,"[\\|/]") %>%
510 .[[1]] %>% 510 .[[1]] %>%
511 .[length(.)] %>% 511 .[length(.)] %>%
512 grepl("soft",.) 512 grepl("soft",.)
513 513
514 ##Changing row names and column names: 514 ##Changing row names and column names:
515 ALZWORD <- t(alzword) 515 ALZWORD <- t(alzword)
516 rownames(ALZWORD)=NULL 516 rownames(ALZWORD)=NULL
517 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE) 517 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE)
518 ALZWORD <- chngrownm(ALZWORD)[-1,] 518 ALZWORD <- chngrownm(ALZWORD)[-1,]
519 ALZWORD <- ALZWORD%>% 519 ALZWORD <- ALZWORD%>%
520 as.data.frame()%>% 520 as.data.frame()%>%
521 dplyr::select(-starts_with("col")) 521 dplyr::select(-starts_with("col"))
522 522
523 ##Reorganizing information within the columns and final clinical data 523 ##Reorganizing information within the columns and final clinical data
524 ALZWORDF <- cinfo(ALZWORD) 524 ALZWORDF <- cinfo(ALZWORD)
525 525
526 526
527 #Working with Actual Data part of file 527 #Working with Actual Data part of file
528 alzdat <- alz %>% 528 alzdat <- alz %>%
529 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1) 529 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1)
530 ALZDAT <- t(alzdat[,-1]) 530 ALZDAT <- t(alzdat[,-1])
531 rownames(ALZDAT)=NULL 531 rownames(ALZDAT)=NULL
532 532
533 ##Is there a clean version of the GPL file available? 533 ##Is there a clean version of the GPL file available?
534 gplnum <- strsplit(genena,"[\\|/]") %>% 534 gplnum <- strsplit(genena,"[\\|/]") %>%
535 .[[1]] %>% 535 .[[1]] %>%
536 .[length(.)] %>% 536 .[length(.)] %>%
537 gsub("\\D","",.) 537 gsub("\\D","",.)
538 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files())) 538 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files()))
539 if(clfileex >= 1){ 539 if(clfileex >= 1){
540 #use the clean version 540 #use the clean version
541 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>% 541 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>%
542 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!") 542 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!")
543 543
544 } 544 }
545 else if(clfileex == 0){ 545 else if(clfileex == 0){
546 ##Lets Create a clean version 546 ##Lets Create a clean version
547 547
548 ##Gene ID to Gene Name 548 ##Gene ID to Gene Name
549 if(soft == TRUE){ 549 if(soft == TRUE){
550 #Check to see if there is already a file containing information on soft files 550 #Check to see if there is already a file containing information on soft files
551 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files())) 551 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files()))
552 if(fileex == 1){ 552 if(fileex == 1){
553 #Check to see if this GPL soft file has been used before 553 #Check to see if this GPL soft file has been used before
554 IDF <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 554 IDF <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
555 .$GPL_FILE_NUM%>% 555 .$GPL_FILE_NUM%>%
556 grepl(gplnum,.) %>% 556 grepl(gplnum,.) %>%
557 sum() 557 sum()
558 if(IDF == 1){ 558 if(IDF == 1){
559 IDLOCAL <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 559 IDLOCAL <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
560 .$GPL_FILE_NUM%>% 560 .$GPL_FILE_NUM%>%
561 grep(gplnum,.) 561 grep(gplnum,.)
562 idlocgpl <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>% 562 idlocgpl <- read_delim("GPL_ID_LOC.txt",delim = "\t",col_names = TRUE) %>%
563 .$LOC_ID %>% 563 .$LOC_ID %>%
564 .[IDLOCAL] 564 .[IDLOCAL]
565 geneIDNam <- genena %>% 565 geneIDNam <- genena %>%
566 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>% 566 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>%
567 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 567 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
568 } 568 }
569 else if(IDF == 0){ 569 else if(IDF == 0){
570 #No information on this particular GPL file 570 #No information on this particular GPL file
571 idLOCGPL <- genena %>% 571 idLOCGPL <- genena %>%
572 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 572 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
573 t(.) %>% 573 t(.) %>%
574 grep("^ID\\s*$",.) %>% 574 grep("^ID\\s*$",.) %>%
575 -1 575 -1
576 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>% 576 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>%
577 cat(file="GPL_ID_LOC.txt",sep = "\t", fill = TRUE, append = TRUE) 577 cat(file="GPL_ID_LOC.txt",sep = "\t", fill = TRUE, append = TRUE)
578 geneIDNam <- genena %>% 578 geneIDNam <- genena %>%
579 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 579 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
580 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 580 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
581 } 581 }
582 } 582 }
583 else if(fileex == 0){ 583 else if(fileex == 0){
584 #We must create a file that we can access for later use 584 #We must create a file that we can access for later use
585 idLOCGPL <- genena %>% 585 idLOCGPL <- genena %>%
586 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 586 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
587 t(.) %>% 587 t(.) %>%
588 grep("^ID\\s*$",.) %>% 588 grep("^ID\\s*$",.) %>%
589 -1 589 -1
590 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL)) 590 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL))
591 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID") 591 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID")
592 write.table(Firstval,file = "GPL_ID_LOC.txt", sep = "\t",row.names = FALSE, col.names = TRUE) 592 write.table(Firstval,file = "GPL_ID_LOC.txt", sep = "\t",row.names = FALSE, col.names = TRUE)
593 geneIDNam <- genena %>% 593 geneIDNam <- genena %>%
594 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 594 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
595 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 595 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
596 } 596 }
597 } 597 }
598 else if(soft == FALSE){ 598 else if(soft == FALSE){
599 geneIDNam <- genena %>% 599 geneIDNam <- genena %>%
600 read_delim(delim="\t",comment = "#")%>% 600 read_delim(delim="\t",comment = "#")%>%
601 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 601 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
602 } 602 }
603 603
604 ##Labeling the gene IDs without names 604 ##Labeling the gene IDs without names
605 geneIDNam <- NAFIXING(geneIDNam) 605 geneIDNam <- NAFIXING(geneIDNam)
606 606
607 ##remove the whitespace 607 ##remove the whitespace
608 geneIDNam <- t(rbind(str_trim(t(geneIDNam)[1,]),str_trim(t(geneIDNam)[2,]))) 608 geneIDNam <- t(rbind(str_trim(t(geneIDNam)[1,]),str_trim(t(geneIDNam)[2,])))
609 609
610 ##Here is the clean version 610 ##Here is the clean version
611 write.table(geneIDNam,file = paste0("Clean_GPL",gplnum,".txt"),sep = "\t",row.names = FALSE, col.names = FALSE) 611 write.table(geneIDNam,file = paste0("Clean_GPL",gplnum,".txt"),sep = "\t",row.names = FALSE, col.names = FALSE)
612 } 612 }
613 613
614 614
615 615
616 ##Changing the gene ID to gene name 616 ##Changing the gene ID to gene name
617 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat)) 617 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat))
618 colnames(ALZDAT) = ALZDAT1[1,] 618 colnames(ALZDAT) = ALZDAT1[1,]
619 619
620 620
621 ##Adjusting the column names aka the gene names 621 ##Adjusting the column names aka the gene names
622 colnames(ALZDAT) <- gcnames(ALZDAT) 622 colnames(ALZDAT) <- gcnames(ALZDAT)
623 623
624 624
625 #Full RAW Data 625 #Full RAW Data
626 Fullalzdwr <- ALZDAT %>% 626 Fullalzdwr <- ALZDAT %>%
627 as.data.frame() %>% 627 as.data.frame() %>%
628 cbind(ALZWORDF,.) 628 cbind(ALZWORDF,.)
629 629
630 #Raw file is output 630 #Raw file is output
631 nfnaex <- strsplit(alz,"[\\]") %>% 631 nfnaex <- strsplit(alz,"[\\]") %>%
632 .[[1]] %>% 632 .[[1]] %>%
633 .[length(.)] %>% 633 .[length(.)] %>%
634 gsub("\\D","",.) %>% 634 gsub("\\D","",.) %>%
635 c("GSE",.,"aftexcel.txt") %>% 635 c("GSE",.,"aftexcel.txt") %>%
636 paste(collapse = "") 636 paste(collapse = "")
637 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t") 637 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t")
638 638
639 639
640 640
641 #Now for the discretization part 641 #Now for the discretization part
642 ##get the wordy part again 642 ##get the wordy part again
643 rawword <- t(ALZWORDF) 643 rawword <- t(ALZWORDF)
644 644
645 ##where is ID_REF located 645 ##where is ID_REF located
646 hereim <- grep("ID_REF",rownames(rawword)) 646 hereim <- grep("ID_REF",rownames(rawword))
647 647
648 ##Subject Names GSM... 648 ##Subject Names GSM...
649 subjnam <- rawword[hereim,] 649 subjnam <- rawword[hereim,]
650 650
651 ##Getting the names for the rows 651 ##Getting the names for the rows
652 namedarows <- rownames(rawword)[-hereim] %>% 652 namedarows <- rownames(rawword)[-hereim] %>%
653 as.data.frame() 653 as.data.frame()
654 RAWWORD <- rawword[-hereim,] %>% 654 RAWWORD <- rawword[-hereim,] %>%
655 as.data.frame() %>% 655 as.data.frame() %>%
656 bind_cols(namedarows,.) 656 bind_cols(namedarows,.)
657 z <- 1 657 z <- 1
658 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE) 658 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE)
659 for(z in 1:dim(RAWWORD)[1]){ 659 for(z in 1:dim(RAWWORD)[1]){
660 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,]))) 660 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,])))
661 z <- z + 1 661 z <- z + 1
662 } 662 }
663 663
664 colnames(naroww) <- "ROW_NAs" 664 colnames(naroww) <- "ROW_NAs"
665 RAWWORD <- bind_cols(RAWWORD,naroww) 665 RAWWORD <- bind_cols(RAWWORD,naroww)
666 666
667 667
668 roALZna <- t(ALZDAT) %>% 668 roALZna <- t(ALZDAT) %>%
669 rownames(.) %>% 669 rownames(.) %>%
670 as.data.frame(.) 670 as.data.frame(.)
671 colnames(roALZna) <- "ID_REF" 671 colnames(roALZna) <- "ID_REF"
672 672
673 RAWDAT <- t(ALZDAT) %>% 673 RAWDAT <- t(ALZDAT) %>%
674 as.data.frame(.) 674 as.data.frame(.)
675 colnames(RAWDAT) <- NULL 675 colnames(RAWDAT) <- NULL
676 rownames(RAWDAT) <- NULL 676 rownames(RAWDAT) <- NULL
677 677
678 RAWDAT2 <- RAWDAT %>% 678 RAWDAT2 <- RAWDAT %>%
679 cbind(roALZna,.) %>% 679 cbind(roALZna,.) %>%
680 dplyr::arrange(.,ID_REF) 680 dplyr::arrange(.,ID_REF)
681 681
682 ##Editing the file for R processing 682 ##Editing the file for R processing
683 RAWDATID <- RAWDAT2[,1] %>% 683 RAWDATID <- RAWDAT2[,1] %>%
684 as.matrix(.) 684 as.matrix(.)
685 685
686 RAWDATNUM <- RAWDAT2[,-1] %>% 686 RAWDATNUM <- RAWDAT2[,-1] %>%
687 mapply(.,FUN = as.numeric) %>% 687 mapply(.,FUN = as.numeric) %>%
688 t(.) 688 t(.)
689 689
690 ##Consolidating genes with the same name 690 ##Consolidating genes with the same name
691 ###create empty matrix of size equal to tabRDATID 691 ###create empty matrix of size equal to tabRDATID
692 tabRDATID <- table(RAWDATID) 692 tabRDATID <- table(RAWDATID)
693 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID)) 693 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID))
694 j <- 1 694 j <- 1
695 for(j in 1:length(tabRDATID)){ 695 for(j in 1:length(tabRDATID)){
696 ##Putting the ones without duplicates in their new homes 696 ##Putting the ones without duplicates in their new homes
697 if(tabRDATID[j] == 1){ 697 if(tabRDATID[j] == 1){
698 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])] 698 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])]
699 } 699 }
700 ##Averaging duplicates and putting them in their new homes 700 ##Averaging duplicates and putting them in their new homes
701 else if(tabRDATID[j] > 1){ 701 else if(tabRDATID[j] > 1){
702 NuRDATN[,j] <- rowMeans(RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])],na.rm = TRUE) 702 NuRDATN[,j] <- rowMeans(RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])],na.rm = TRUE)
703 } 703 }
704 j <- j + 1 704 j <- j + 1
705 } 705 }
706 706
707 ##Scaling the Data 707 ##Scaling the Data
708 scrawdat <- NuRDATN%>% 708 scrawdat <- NuRDATN%>%
709 scale() 709 scale()
710 attr(scrawdat,"scaled:center") <- NULL 710 attr(scrawdat,"scaled:center") <- NULL
711 attr(scrawdat,"scaled:scale") <- NULL 711 attr(scrawdat,"scaled:scale") <- NULL
712 colnames(scrawdat) <- rownames(tabRDATID) 712 colnames(scrawdat) <- rownames(tabRDATID)
713 713
714 ##Discretized the Data 714 ##Discretized the Data
715 dialzdat <- scrawdat %>% 715 dialzdat <- scrawdat %>%
716 dndat(.) %>% 716 dndat(.) %>%
717 t()%>% 717 t()%>%
718 as.data.frame(.) 718 as.data.frame(.)
719 colnames(dialzdat) <- rownames(RAWDATNUM) 719 colnames(dialzdat) <- rownames(RAWDATNUM)
720 720
721 ##setting "ID_REF" as a new variable 721 ##setting "ID_REF" as a new variable
722 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1)) 722 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1))
723 colnames(geneNAM) <- "ID_REF" 723 colnames(geneNAM) <- "ID_REF"
724 rownames(dialzdat) <- NULL 724 rownames(dialzdat) <- NULL
725 dialzdat <-bind_cols(geneNAM,dialzdat) 725 dialzdat <-bind_cols(geneNAM,dialzdat)
726 726
727 ##NAs in a column 727 ##NAs in a column
728 x <- 2 728 x <- 2
729 nacol <- as.data.frame(t(rep(0,dim(dialzdat)[2])),stringsAsFactors = FALSE) 729 nacol <- as.data.frame(t(rep(0,dim(dialzdat)[2])),stringsAsFactors = FALSE)
730 nacol[1,1] = "COL_NAs" 730 nacol[1,1] = "COL_NAs"
731 for(x in 2:dim(dialzdat)[2]){ 731 for(x in 2:dim(dialzdat)[2]){
732 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x]))) 732 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x])))
733 x <- x + 1 733 x <- x + 1
734 } 734 }
735 colnames(nacol) <- colnames(dialzdat) 735 colnames(nacol) <- colnames(dialzdat)
736 dialzdat <- bind_rows(dialzdat,nacol) 736 dialzdat <- bind_rows(dialzdat,nacol)
737 737
738 ##NAs in a row 738 ##NAs in a row
739 y <- 1 739 y <- 1
740 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE) 740 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE)
741 for(y in 1:dim(dialzdat)[1]){ 741 for(y in 1:dim(dialzdat)[1]){
742 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,]))) 742 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,])))
743 y <- y + 1 743 y <- y + 1
744 } 744 }
745 colnames(narowd) <- "ROW_NAs" 745 colnames(narowd) <- "ROW_NAs"
746 dialzdat <- bind_cols(dialzdat,narowd) 746 dialzdat <- bind_cols(dialzdat,narowd)
747 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam 747 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam
748 colnames(RAWWORD) <- colnames(dialzdat) 748 colnames(RAWWORD) <- colnames(dialzdat)
749 ##converting to character so that the clinical can be brought together with discrete data 749 ##converting to character so that the clinical can be brought together with discrete data
750 k <- 2 750 k <- 2
751 for(k in 2:dim(dialzdat)[2]-1){ 751 for(k in 2:dim(dialzdat)[2]-1){
752 dialzdat[,k] <- as.character(dialzdat[,k]) 752 dialzdat[,k] <- as.character(dialzdat[,k])
753 k <- k + 1 753 k <- k + 1
754 } 754 }
755 #The End the full data 755 #The End the full data
756 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat) 756 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat)
757 757
758 #Produces Discrete file 758 #Produces Discrete file
759 nfnaex2 <- strsplit(alz,"[\\|/]") %>% 759 nfnaex2 <- strsplit(alz,"[\\|/]") %>%
760 .[[1]] %>% 760 .[[1]] %>%
761 .[length(.)] %>% 761 .[length(.)] %>%
762 gsub("\\D","",.) %>% 762 gsub("\\D","",.) %>%
763 c("GSE",.,"dscrt.txt") %>% 763 c("GSE",.,"dscrt.txt") %>%
764 paste(collapse = "") 764 paste(collapse = "")
765 write.table(Dscrtalzdw, file = nfnaex2, sep = "\t",col.names = TRUE,row.names = FALSE) 765 write.table(Dscrtalzdw, file = nfnaex2, sep = "\t",col.names = TRUE,row.names = FALSE)
766 766
767 767
768 n <- n + 1 768 n <- n + 1
769 } 769 }
770 } 770 }
771 } 771 }
772 } 772 }
773 #The Rest of this code will be used every time you want to change a data set 773 #The Rest of this code will be used every time you want to change a data set
774 THEFT() 774 THEFT()