Efrain Gonzalez / Cleaning and Fixing Data with R

Download as
Browse Code ยป

Commit 69ac507843530d76722471db7ddbf5f213a11aee

Authored by Efrain Gonzalez
1 parent ae780ea650
Exists in master

Don't use this code yet 

Updated to decrease time of search for cgeneID
Showing 1 changed file with 29 additions and 11 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 else if(grepl("postmorteminterval|PMI|pmi",mat[2,e])==TRUE){ 41 else if(grepl("postmorteminterval|PMI|pmi",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 else if(grepl("age|Age|AGE",mat[2,e])==TRUE){ 45 else 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 else if(grepl("braak|b&b",mat[2,e])==TRUE){ 49 else 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 else if(grepl("group|disease|control|AD|normal|diagnosis|Alzheimer|Control|Normal",mat[2,e])==TRUE){ 53 else 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 cgeneID2 <- function(GeneName,DATA){
108 colGene <- dim(GeneName)[2] 108 nj <- t(GeneName)
109 j <- 1 109 nq <- t(DATA)
110 for(j in 1:colGene){ 110 colGene <- dim(nj)[2]
111 chngsreq <- grep(paste0("^",GeneName[1,j],"$"),DATA[1,]) 111 colDATA <- dim(nq)[2]
112 if(is.na(sum(chngsreq))==FALSE){ 112 j <- 1
113 if(sum(chngsreq) > 0){ 113 for(j in 1:colDATA){
114 DATA[1,chngsreq] <- gsub(paste0("^",GeneName[1,j]),GeneName[2,j],DATA[1,chngsreq]) 114 #where is that gene id located within the GPL file
115 chngreq <- grep(paste0("^",nq[1,j],"$"),nj[1,])
116 if(is.na(sum(chngreq))==FALSE){
117 if(sum(chngreq) > 0){
118 nq[1,j] <- gsub(paste0("^",nq[1,j],"$"),nj[2,chngreq],nq[1,j])
119 }
115 } 120 }
121 j <- j + 1
116 } 122 }
117 j = j+1 123 nq
118 }
119 DATA
120 } 124 }
125 #cgeneID <- function(GeneName,DATA){
126 # colGene <- dim(GeneName)[2]
127 # j <- 1
128 # for(j in 1:colGene){
129 # chngsreq <- grep(paste0("^",GeneName[1,j],"$"),DATA[1,])
130 # if(is.na(sum(chngsreq))==FALSE){
131 # if(sum(chngsreq) > 0){
132 # DATA[1,chngsreq] <- gsub(paste0("^",GeneName[1,j]),GeneName[2,j],DATA[1,chngsreq])
133 # }
134 # }
135 # j = j+1
136 # }
137 # DATA
138 #}
121 139
122 #5#Function for adjusting the gene names 140 #5#Function for adjusting the gene names
123 gcnames <- function(DiData,usecol=1){ 141 gcnames <- function(DiData,usecol=1){
124 nuruns <- dim(DiData)[2] 142 nuruns <- dim(DiData)[2]
125 i = 1 143 i = 1
126 nwnam <- rep("0",length.out=nuruns) 144 nwnam <- rep("0",length.out=nuruns)
127 for(i in 1:nuruns){ 145 for(i in 1:nuruns){
128 if(length(strsplit(colnames(DiData)[i],"///")[[1]]) >= usecol){ 146 if(length(strsplit(colnames(DiData)[i],"///")[[1]]) >= usecol){
129 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][usecol]) 147 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][usecol])
130 } else{ 148 } else{
131 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][1]) 149 nwnam[i]=str_trim(strsplit(colnames(DiData)[i],"///")[[1]][1])
132 } 150 }
133 151
134 } 152 }
135 nwnam 153 nwnam
136 154
137 } 155 }
138 156
139 #6# Function for discretizing the data 157 #6# Function for discretizing the data
140 dndat <- function(NDATA){ 158 dndat <- function(NDATA){
141 rownd <- dim(NDATA)[1] 159 rownd <- dim(NDATA)[1]
142 colnd <- dim(NDATA)[2] 160 colnd <- dim(NDATA)[2]
143 DDATA <- matrix(0,nrow=rownd,ncol=colnd) 161 DDATA <- matrix(0,nrow=rownd,ncol=colnd)
144 colnames(DDATA) <- colnames(NDATA) 162 colnames(DDATA) <- colnames(NDATA)
145 i <- 1 163 i <- 1
146 for(i in 1:rownd){ 164 for(i in 1:rownd){
147 j <- 1 165 j <- 1
148 for(j in 1:colnd){ 166 for(j in 1:colnd){
149 if(is.na(NDATA[i,j])==FALSE){ 167 if(is.na(NDATA[i,j])==FALSE){
150 168
151 if(NDATA[i,j] < -1){ 169 if(NDATA[i,j] < -1){
152 DDATA[i,j]=0L 170 DDATA[i,j]=0L
153 } 171 }
154 if(NDATA[i,j] > 1){ 172 if(NDATA[i,j] > 1){
155 DDATA[i,j]=2L 173 DDATA[i,j]=2L
156 } 174 }
157 if(-1 <= NDATA[i,j] && NDATA[i,j] < 1){ 175 if(-1 <= NDATA[i,j] && NDATA[i,j] < 1){
158 DDATA[i,j]=1L 176 DDATA[i,j]=1L
159 } 177 }
160 } else{ 178 } else{
161 DDATA[i,j] = NDATA[i,j] 179 DDATA[i,j] = NDATA[i,j]
162 } 180 }
163 j = j + 1 181 j = j + 1
164 } 182 }
165 i = i + 1 183 i = i + 1
166 } 184 }
167 DDATA 185 DDATA
168 } 186 }
169 187
170 188
171 #MajorFunction#This is the function that does everything else 189 #MajorFunction#This is the function that does everything else
172 THEFT <- function(){ 190 THEFT <- function(){
173 #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
174 wd <- getwd() 192 wd <- getwd()
175 #list.files() 193 #list.files()
176 #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?")
177 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)
178 GSEfileloc <- grep("^GSE.+\\.txt\\.gz$",list.files()) 196 GSEfileloc <- grep("^GSE.+\\.txt\\.gz$",list.files())
179 GSEfloc <- list.files()[GSEfileloc] 197 GSEfloc <- list.files()[GSEfileloc]
180 #ALL DATA FILES WILL BE CLEANED 198 #ALL DATA FILES WILL BE CLEANED
181 if(numDAT == 1){ 199 if(numDAT == 1){
182 #indexing the data files 200 #indexing the data files
183 n <- 1 201 n <- 1
184 for(n in 1: length(GSEfloc)){ 202 for(n in 1: length(GSEfloc)){
185 alz <- GSEfloc[n] 203 alz <- GSEfloc[n]
186 204
187 #Working with the wordy part of the document 205 #Working with the wordy part of the document
188 alzword <- alz %>% 206 alzword <- alz %>%
189 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>% 207 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>%
190 filter(grepl("!Sample",X1))%>% 208 filter(grepl("!Sample",X1))%>%
191 filter(!grepl("!Sample_contact",X1)) 209 filter(!grepl("!Sample_contact",X1))
192 210
193 #Getting the GPL file 211 #Getting the GPL file
194 genena <- grep("_platform_id",alzword$X1) %>% 212 genena <- grep("_platform_id",alzword$X1) %>%
195 alzword$X2[.] %>% 213 alzword$X2[.] %>%
196 str_trim(.) %>% 214 str_trim(.) %>%
197 paste0("^",.,"\\D") %>% 215 paste0("^",.,"\\D") %>%
198 grep(.,list.files()) %>% 216 grep(.,list.files()) %>%
199 list.files()[.] 217 list.files()[.]
200 218
201 #Find out if it is a soft GPL file or not 219 #Find out if it is a soft GPL file or not
202 soft <- strsplit(genena,"[\\|/]") %>% 220 soft <- strsplit(genena,"[\\|/]") %>%
203 .[[1]] %>% 221 .[[1]] %>%
204 .[length(.)] %>% 222 .[length(.)] %>%
205 grepl("soft",.) 223 grepl("soft",.)
206 224
207 ##Changing row names and column names: 225 ##Changing row names and column names:
208 ALZWORD <- t(alzword) 226 ALZWORD <- t(alzword)
209 rownames(ALZWORD)=NULL 227 rownames(ALZWORD)=NULL
210 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE) 228 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE)
211 ALZWORD <- chngrownm(ALZWORD)[-1,] 229 ALZWORD <- chngrownm(ALZWORD)[-1,]
212 ALZWORD <- ALZWORD%>% 230 ALZWORD <- ALZWORD%>%
213 as.data.frame()%>% 231 as.data.frame()%>%
214 dplyr::select(-starts_with("col")) 232 dplyr::select(-starts_with("col"))
215 233
216 ##Reorganizing information within the columns and final clinical data 234 ##Reorganizing information within the columns and final clinical data
217 ALZWORDF <- cinfo(ALZWORD) 235 ALZWORDF <- cinfo(ALZWORD)
218 236
219 237
220 #Working with Actual Data part of file 238 #Working with Actual Data part of file
221 alzdat <- alz %>% 239 alzdat <- alz %>%
222 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1) 240 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1)
223 ALZDAT <- t(alzdat[,-1]) 241 ALZDAT <- t(alzdat[,-1])
224 rownames(ALZDAT)=NULL 242 rownames(ALZDAT)=NULL
225 243
226 ##Is there a clean version of the GPL file available? 244 ##Is there a clean version of the GPL file available?
227 gplnum <- strsplit(genena,"[\\|/]") %>% 245 gplnum <- strsplit(genena,"[\\|/]") %>%
228 .[[1]] %>% 246 .[[1]] %>%
229 .[length(.)] %>% 247 .[length(.)] %>%
230 gsub("\\D","",.) 248 gsub("\\D","",.)
231 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files())) 249 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files()))
232 if(clfileex >= 1){ 250 if(clfileex >= 1){
233 #use the clean version 251 #use the clean version
234 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>% 252 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>%
235 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!") 253 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!")
236 254
237 } 255 }
238 else if(clfileex == 0){ 256 else if(clfileex == 0){
239 ##Lets Create a clean version 257 ##Lets Create a clean version
240 258
241 ##Gene ID to Gene Name 259 ##Gene ID to Gene Name
242 if(soft == TRUE){ 260 if(soft == TRUE){
243 #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
244 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files())) 262 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files()))
245 if(fileex == 1){ 263 if(fileex == 1){
246 #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
247 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) %>%
248 .$GPL_FILE_NUM%>% 266 .$GPL_FILE_NUM%>%
249 grepl(gplnum,.) %>% 267 grepl(gplnum,.) %>%
250 sum() 268 sum()
251 if(IDF == 1){ 269 if(IDF == 1){
252 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) %>%
253 .$GPL_FILE_NUM%>% 271 .$GPL_FILE_NUM%>%
254 grep(gplnum,.) 272 grep(gplnum,.)
255 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) %>%
256 .$LOC_ID %>% 274 .$LOC_ID %>%
257 .[IDLOCAL] 275 .[IDLOCAL]
258 geneIDNam <- genena %>% 276 geneIDNam <- genena %>%
259 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>% 277 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>%
260 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 278 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
261 } 279 }
262 else if(IDF == 0){ 280 else if(IDF == 0){
263 #No information on this particular GPL file 281 #No information on this particular GPL file
264 idLOCGPL <- genena %>% 282 idLOCGPL <- genena %>%
265 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 283 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
266 t(.) %>% 284 t(.) %>%
267 grep("^ID\\s*$",.) %>% 285 grep("^ID\\s*$",.) %>%
268 -1 286 -1
269 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>% 287 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>%
270 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)
271 geneIDNam <- genena %>% 289 geneIDNam <- genena %>%
272 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 290 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
273 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 291 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
274 } 292 }
275 } 293 }
276 else if(fileex == 0){ 294 else if(fileex == 0){
277 #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
278 idLOCGPL <- genena %>% 296 idLOCGPL <- genena %>%
279 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 297 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
280 t(.) %>% 298 t(.) %>%
281 grep("^ID\\s*$",.) %>% 299 grep("^ID\\s*$",.) %>%
282 -1 300 -1
283 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL)) 301 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL))
284 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID") 302 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID")
285 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)
286 geneIDNam <- genena %>% 304 geneIDNam <- genena %>%
287 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 305 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
288 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 306 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
289 } 307 }
290 } 308 }
291 else if(soft == FALSE){ 309 else if(soft == FALSE){
292 geneIDNam <- genena %>% 310 geneIDNam <- genena %>%
293 read_delim(delim="\t",comment = "#")%>% 311 read_delim(delim="\t",comment = "#")%>%
294 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 312 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
295 } 313 }
296 314
297 ##Labeling the gene IDs without names 315 ##Labeling the gene IDs without names
298 geneIDNam <- NAFIXING(geneIDNam) 316 geneIDNam <- NAFIXING(geneIDNam)
299 317
300 ##remove the whitespace 318 ##remove the whitespace
301 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,])))
302 320
303 ##Here is the clean version 321 ##Here is the clean version
304 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)
305 } 323 }
306 324
307 325
308 326
309 ##Changing the gene ID to gene name 327 ##Changing the gene ID to gene name
310 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat)) 328 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat))
311 colnames(ALZDAT) = ALZDAT1[1,] 329 colnames(ALZDAT) = ALZDAT1[1,]
312 330
313 331
314 ##Adjusting the column names aka the gene names 332 ##Adjusting the column names aka the gene names
315 colnames(ALZDAT) <- gcnames(ALZDAT) 333 colnames(ALZDAT) <- gcnames(ALZDAT)
316 334
317 335
318 #Full RAW Data 336 #Full RAW Data
319 Fullalzdwr <- ALZDAT %>% 337 Fullalzdwr <- ALZDAT %>%
320 as.data.frame() %>% 338 as.data.frame() %>%
321 cbind(ALZWORDF,.) 339 cbind(ALZWORDF,.)
322 340
323 #Raw file is output 341 #Raw file is output
324 nfnaex <- strsplit(alz,"[\\]") %>% 342 nfnaex <- strsplit(alz,"[\\]") %>%
325 .[[1]] %>% 343 .[[1]] %>%
326 .[length(.)] %>% 344 .[length(.)] %>%
327 gsub("\\D","",.) %>% 345 gsub("\\D","",.) %>%
328 c("GSE",.,"aftexcel.txt") %>% 346 c("GSE",.,"aftexcel.txt") %>%
329 paste(collapse = "") 347 paste(collapse = "")
330 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t") 348 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t")
331 349
332 350
333 351
334 #Now for the discretization part 352 #Now for the discretization part
335 ##get the wordy part again 353 ##get the wordy part again
336 rawword <- t(ALZWORDF) 354 rawword <- t(ALZWORDF)
337 355
338 ##where is ID_REF located 356 ##where is ID_REF located
339 hereim <- grep("ID_REF",rownames(rawword)) 357 hereim <- grep("ID_REF",rownames(rawword))
340 358
341 ##Subject Names GSM... 359 ##Subject Names GSM...
342 subjnam <- rawword[hereim,] 360 subjnam <- rawword[hereim,]
343 361
344 ##Getting the names for the rows 362 ##Getting the names for the rows
345 namedarows <- rownames(rawword)[-hereim] %>% 363 namedarows <- rownames(rawword)[-hereim] %>%
346 as.data.frame() 364 as.data.frame()
347 RAWWORD <- rawword[-hereim,] %>% 365 RAWWORD <- rawword[-hereim,] %>%
348 as.data.frame() %>% 366 as.data.frame() %>%
349 bind_cols(namedarows,.) 367 bind_cols(namedarows,.)
350 z <- 1 368 z <- 1
351 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE) 369 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE)
352 for(z in 1:dim(RAWWORD)[1]){ 370 for(z in 1:dim(RAWWORD)[1]){
353 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,]))) 371 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,])))
354 z <- z + 1 372 z <- z + 1
355 } 373 }
356 374
357 colnames(naroww) <- "ROW_NAs" 375 colnames(naroww) <- "ROW_NAs"
358 RAWWORD <- bind_cols(RAWWORD,naroww) 376 RAWWORD <- bind_cols(RAWWORD,naroww)
359 377
360 378
361 roALZna <- t(ALZDAT) %>% 379 roALZna <- t(ALZDAT) %>%
362 rownames(.) %>% 380 rownames(.) %>%
363 as.data.frame(.) 381 as.data.frame(.)
364 colnames(roALZna) <- "ID_REF" 382 colnames(roALZna) <- "ID_REF"
365 383
366 RAWDAT <- t(ALZDAT) %>% 384 RAWDAT <- t(ALZDAT) %>%
367 as.data.frame(.) 385 as.data.frame(.)
368 colnames(RAWDAT) <- NULL 386 colnames(RAWDAT) <- NULL
369 rownames(RAWDAT) <- NULL 387 rownames(RAWDAT) <- NULL
370 388
371 RAWDAT2 <- RAWDAT %>% 389 RAWDAT2 <- RAWDAT %>%
372 cbind(roALZna,.) %>% 390 cbind(roALZna,.) %>%
373 dplyr::arrange(.,ID_REF) 391 dplyr::arrange(.,ID_REF)
374 392
375 ##Editing the file for R processing 393 ##Editing the file for R processing
376 RAWDATID <- RAWDAT2[,1] %>% 394 RAWDATID <- RAWDAT2[,1] %>%
377 as.matrix(.) 395 as.matrix(.)
378 396
379 RAWDATNUM <- RAWDAT2[,-1] %>% 397 RAWDATNUM <- RAWDAT2[,-1] %>%
380 mapply(.,FUN = as.numeric) %>% 398 mapply(.,FUN = as.numeric) %>%
381 t(.) 399 t(.)
382 400
383 ##Consolidating genes with the same name 401 ##Consolidating genes with the same name
384 ###create empty matrix of size equal to tabRDATID 402 ###create empty matrix of size equal to tabRDATID
385 tabRDATID <- table(RAWDATID) 403 tabRDATID <- table(RAWDATID)
386 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID)) 404 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID))
387 j <- 1 405 j <- 1
388 for(j in 1:length(tabRDATID)){ 406 for(j in 1:length(tabRDATID)){
389 ##Putting the ones without duplicates in their new homes 407 ##Putting the ones without duplicates in their new homes
390 if(tabRDATID[j] == 1){ 408 if(tabRDATID[j] == 1){
391 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])] 409 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])]
392 } 410 }
393 ##Averaging duplicates and putting them in their new homes 411 ##Averaging duplicates and putting them in their new homes
394 else if(tabRDATID[j] > 1){ 412 else if(tabRDATID[j] > 1){
395 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)
396 } 414 }
397 j <- j + 1 415 j <- j + 1
398 } 416 }
399 417
400 ##Scaling the Data 418 ##Scaling the Data
401 scrawdat <- NuRDATN%>% 419 scrawdat <- NuRDATN%>%
402 scale() 420 scale()
403 attr(scrawdat,"scaled:center") <- NULL 421 attr(scrawdat,"scaled:center") <- NULL
404 attr(scrawdat,"scaled:scale") <- NULL 422 attr(scrawdat,"scaled:scale") <- NULL
405 colnames(scrawdat) <- rownames(tabRDATID) 423 colnames(scrawdat) <- rownames(tabRDATID)
406 424
407 ##Discretized the Data 425 ##Discretized the Data
408 dialzdat <- scrawdat %>% 426 dialzdat <- scrawdat %>%
409 dndat(.) %>% 427 dndat(.) %>%
410 t()%>% 428 t()%>%
411 as.data.frame(.) 429 as.data.frame(.)
412 colnames(dialzdat) <- rownames(RAWDATNUM) 430 colnames(dialzdat) <- rownames(RAWDATNUM)
413 431
414 ##setting "ID_REF" as a new variable 432 ##setting "ID_REF" as a new variable
415 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1)) 433 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1))
416 colnames(geneNAM) <- "ID_REF" 434 colnames(geneNAM) <- "ID_REF"
417 rownames(dialzdat) <- NULL 435 rownames(dialzdat) <- NULL
418 dialzdat <-bind_cols(geneNAM,dialzdat) 436 dialzdat <-bind_cols(geneNAM,dialzdat)
419 437
420 ##NAs in a column 438 ##NAs in a column
421 x <- 2 439 x <- 2
422 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)
423 nacol[1,1] = "COL_NAs" 441 nacol[1,1] = "COL_NAs"
424 for(x in 2:dim(dialzdat)[2]){ 442 for(x in 2:dim(dialzdat)[2]){
425 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x]))) 443 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x])))
426 x <- x + 1 444 x <- x + 1
427 } 445 }
428 colnames(nacol) <- colnames(dialzdat) 446 colnames(nacol) <- colnames(dialzdat)
429 dialzdat <- bind_rows(dialzdat,nacol) 447 dialzdat <- bind_rows(dialzdat,nacol)
430 448
431 ##NAs in a row 449 ##NAs in a row
432 y <- 1 450 y <- 1
433 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE) 451 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE)
434 for(y in 1:dim(dialzdat)[1]){ 452 for(y in 1:dim(dialzdat)[1]){
435 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,]))) 453 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,])))
436 y <- y + 1 454 y <- y + 1
437 } 455 }
438 colnames(narowd) <- "ROW_NAs" 456 colnames(narowd) <- "ROW_NAs"
439 dialzdat <- bind_cols(dialzdat,narowd) 457 dialzdat <- bind_cols(dialzdat,narowd)
440 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam 458 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam
441 colnames(RAWWORD) <- colnames(dialzdat) 459 colnames(RAWWORD) <- colnames(dialzdat)
442 ##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
443 k <- 2 461 k <- 2
444 for(k in 2:dim(dialzdat)[2]-1){ 462 for(k in 2:dim(dialzdat)[2]-1){
445 dialzdat[,k] <- as.character(dialzdat[,k]) 463 dialzdat[,k] <- as.character(dialzdat[,k])
446 k <- k + 1 464 k <- k + 1
447 } 465 }
448 #The End the full data 466 #The End the full data
449 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat) 467 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat)
450 468
451 #Produces Discrete file 469 #Produces Discrete file
452 nfnaex2 <- strsplit(alz,"[\\|/]") %>% 470 nfnaex2 <- strsplit(alz,"[\\|/]") %>%
453 .[[1]] %>% 471 .[[1]] %>%
454 .[length(.)] %>% 472 .[length(.)] %>%
455 gsub("\\D","",.) %>% 473 gsub("\\D","",.) %>%
456 c("GSE",.,"dscrt.txt") %>% 474 c("GSE",.,"dscrt.txt") %>%
457 paste(collapse = "") 475 paste(collapse = "")
458 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)
459 n <- n +1 477 n <- n +1
460 } 478 }
461 } 479 }
462 480
463 #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
464 else if(numDAT == 2){ 482 else if(numDAT == 2){
465 #All the files you want to analyze 483 #All the files you want to analyze
466 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:")
467 if(length(ANDIS) == 0){ 485 if(length(ANDIS) == 0){
468 #Spit out a warning 486 #Spit out a warning
469 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")
470 } else{ 488 } else{
471 #indexing the data files 489 #indexing the data files
472 n <- 1 490 n <- 1
473 for(n in 1: length(ANDIS)){ 491 for(n in 1: length(ANDIS)){
474 alz <- ANDIS[n] 492 alz <- ANDIS[n]
475 493
476 #Working with the wordy part of the document 494 #Working with the wordy part of the document
477 alzword <- alz %>% 495 alzword <- alz %>%
478 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>% 496 read_delim(delim ="\t",comment = "!Series",col_names = FALSE)%>%
479 filter(grepl("!Sample",X1))%>% 497 filter(grepl("!Sample",X1))%>%
480 filter(!grepl("!Sample_contact",X1)) 498 filter(!grepl("!Sample_contact",X1))
481 499
482 #Getting the GPL file 500 #Getting the GPL file
483 genena <- grep("_platform_id",alzword$X1) %>% 501 genena <- grep("_platform_id",alzword$X1) %>%
484 alzword$X2[.] %>% 502 alzword$X2[.] %>%
485 str_trim(.) %>% 503 str_trim(.) %>%
486 paste0("^",.,"\\D") %>% 504 paste0("^",.,"\\D") %>%
487 grep(.,list.files()) %>% 505 grep(.,list.files()) %>%
488 list.files()[.] 506 list.files()[.]
489 507
490 #Find out if it is a soft GPL file or not 508 #Find out if it is a soft GPL file or not
491 soft <- strsplit(genena,"[\\|/]") %>% 509 soft <- strsplit(genena,"[\\|/]") %>%
492 .[[1]] %>% 510 .[[1]] %>%
493 .[length(.)] %>% 511 .[length(.)] %>%
494 grepl("soft",.) 512 grepl("soft",.)
495 513
496 ##Changing row names and column names: 514 ##Changing row names and column names:
497 ALZWORD <- t(alzword) 515 ALZWORD <- t(alzword)
498 rownames(ALZWORD)=NULL 516 rownames(ALZWORD)=NULL
499 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE) 517 colnames(ALZWORD) <- colnames(ALZWORD,do.NULL=FALSE)
500 ALZWORD <- chngrownm(ALZWORD)[-1,] 518 ALZWORD <- chngrownm(ALZWORD)[-1,]
501 ALZWORD <- ALZWORD%>% 519 ALZWORD <- ALZWORD%>%
502 as.data.frame()%>% 520 as.data.frame()%>%
503 dplyr::select(-starts_with("col")) 521 dplyr::select(-starts_with("col"))
504 522
505 ##Reorganizing information within the columns and final clinical data 523 ##Reorganizing information within the columns and final clinical data
506 ALZWORDF <- cinfo(ALZWORD) 524 ALZWORDF <- cinfo(ALZWORD)
507 525
508 526
509 #Working with Actual Data part of file 527 #Working with Actual Data part of file
510 alzdat <- alz %>% 528 alzdat <- alz %>%
511 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1) 529 read_delim(delim="\t",col_names=TRUE,comment = "!",skip=1)
512 ALZDAT <- t(alzdat[,-1]) 530 ALZDAT <- t(alzdat[,-1])
513 rownames(ALZDAT)=NULL 531 rownames(ALZDAT)=NULL
514 532
515 ##Is there a clean version of the GPL file available? 533 ##Is there a clean version of the GPL file available?
516 gplnum <- strsplit(genena,"[\\|/]") %>% 534 gplnum <- strsplit(genena,"[\\|/]") %>%
517 .[[1]] %>% 535 .[[1]] %>%
518 .[length(.)] %>% 536 .[length(.)] %>%
519 gsub("\\D","",.) 537 gsub("\\D","",.)
520 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files())) 538 clfileex <- sum(grepl(paste0("Clean_GPL",gplnum),list.files()))
521 if(clfileex >= 1){ 539 if(clfileex >= 1){
522 #use the clean version 540 #use the clean version
523 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>% 541 geneIDNam <- paste0("Clean_GPL",gplnum,".txt") %>%
524 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!") 542 read_delim(delim="\t",col_names = c("ID","Symbol"), comment = "!")
525 543
526 } 544 }
527 else if(clfileex == 0){ 545 else if(clfileex == 0){
528 ##Lets Create a clean version 546 ##Lets Create a clean version
529 547
530 ##Gene ID to Gene Name 548 ##Gene ID to Gene Name
531 if(soft == TRUE){ 549 if(soft == TRUE){
532 #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
533 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files())) 551 fileex <- sum(grepl("GPL_ID_LOC.txt",list.files()))
534 if(fileex == 1){ 552 if(fileex == 1){
535 #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
536 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) %>%
537 .$GPL_FILE_NUM%>% 555 .$GPL_FILE_NUM%>%
538 grepl(gplnum,.) %>% 556 grepl(gplnum,.) %>%
539 sum() 557 sum()
540 if(IDF == 1){ 558 if(IDF == 1){
541 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) %>%
542 .$GPL_FILE_NUM%>% 560 .$GPL_FILE_NUM%>%
543 grep(gplnum,.) 561 grep(gplnum,.)
544 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) %>%
545 .$LOC_ID %>% 563 .$LOC_ID %>%
546 .[IDLOCAL] 564 .[IDLOCAL]
547 geneIDNam <- genena %>% 565 geneIDNam <- genena %>%
548 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>% 566 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idlocgpl) %>%
549 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 567 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
550 } 568 }
551 else if(IDF == 0){ 569 else if(IDF == 0){
552 #No information on this particular GPL file 570 #No information on this particular GPL file
553 idLOCGPL <- genena %>% 571 idLOCGPL <- genena %>%
554 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 572 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
555 t(.) %>% 573 t(.) %>%
556 grep("^ID\\s*$",.) %>% 574 grep("^ID\\s*$",.) %>%
557 -1 575 -1
558 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>% 576 cbind(as.integer(gplnum),as.integer(idLOCGPL)) %>%
559 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)
560 geneIDNam <- genena %>% 578 geneIDNam <- genena %>%
561 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 579 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
562 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 580 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
563 } 581 }
564 } 582 }
565 else if(fileex == 0){ 583 else if(fileex == 0){
566 #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
567 idLOCGPL <- genena %>% 585 idLOCGPL <- genena %>%
568 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>% 586 read_delim(delim="\t",col_names = FALSE, comment = "!", n_max = 1000) %>%
569 t(.) %>% 587 t(.) %>%
570 grep("^ID\\s*$",.) %>% 588 grep("^ID\\s*$",.) %>%
571 -1 589 -1
572 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL)) 590 Firstval <- cbind(as.integer(gplnum),as.integer(idLOCGPL))
573 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID") 591 colnames(Firstval) <- c("GPL_FILE_NUM","LOC_ID")
574 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)
575 geneIDNam <- genena %>% 593 geneIDNam <- genena %>%
576 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>% 594 read_delim(delim="\t",col_names = TRUE, comment = "!", skip = idLOCGPL) %>%
577 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 595 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
578 } 596 }
579 } 597 }
580 else if(soft == FALSE){ 598 else if(soft == FALSE){
581 geneIDNam <- genena %>% 599 geneIDNam <- genena %>%
582 read_delim(delim="\t",comment = "#")%>% 600 read_delim(delim="\t",comment = "#")%>%
583 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.))) 601 dplyr::select(.,ID,grep("Symbol|^ORF\\s*$|^gene_assignment\\s*$",colnames(.)))
584 } 602 }
585 603
586 ##Labeling the gene IDs without names 604 ##Labeling the gene IDs without names
587 geneIDNam <- NAFIXING(geneIDNam) 605 geneIDNam <- NAFIXING(geneIDNam)
588 606
589 ##remove the whitespace 607 ##remove the whitespace
590 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,])))
591 609
592 ##Here is the clean version 610 ##Here is the clean version
593 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)
594 } 612 }
595 613
596 614
597 615
598 ##Changing the gene ID to gene name 616 ##Changing the gene ID to gene name
599 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat)) 617 ALZDAT1 <- cgeneID(t(geneIDNam),t(alzdat))
600 colnames(ALZDAT) = ALZDAT1[1,] 618 colnames(ALZDAT) = ALZDAT1[1,]
601 619
602 620
603 ##Adjusting the column names aka the gene names 621 ##Adjusting the column names aka the gene names
604 colnames(ALZDAT) <- gcnames(ALZDAT) 622 colnames(ALZDAT) <- gcnames(ALZDAT)
605 623
606 624
607 #Full RAW Data 625 #Full RAW Data
608 Fullalzdwr <- ALZDAT %>% 626 Fullalzdwr <- ALZDAT %>%
609 as.data.frame() %>% 627 as.data.frame() %>%
610 cbind(ALZWORDF,.) 628 cbind(ALZWORDF,.)
611 629
612 #Raw file is output 630 #Raw file is output
613 nfnaex <- strsplit(alz,"[\\]") %>% 631 nfnaex <- strsplit(alz,"[\\]") %>%
614 .[[1]] %>% 632 .[[1]] %>%
615 .[length(.)] %>% 633 .[length(.)] %>%
616 gsub("\\D","",.) %>% 634 gsub("\\D","",.) %>%
617 c("GSE",.,"aftexcel.txt") %>% 635 c("GSE",.,"aftexcel.txt") %>%
618 paste(collapse = "") 636 paste(collapse = "")
619 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t") 637 write.table(t(Fullalzdwr), file = nfnaex, sep = "\t")
620 638
621 639
622 640
623 #Now for the discretization part 641 #Now for the discretization part
624 ##get the wordy part again 642 ##get the wordy part again
625 rawword <- t(ALZWORDF) 643 rawword <- t(ALZWORDF)
626 644
627 ##where is ID_REF located 645 ##where is ID_REF located
628 hereim <- grep("ID_REF",rownames(rawword)) 646 hereim <- grep("ID_REF",rownames(rawword))
629 647
630 ##Subject Names GSM... 648 ##Subject Names GSM...
631 subjnam <- rawword[hereim,] 649 subjnam <- rawword[hereim,]
632 650
633 ##Getting the names for the rows 651 ##Getting the names for the rows
634 namedarows <- rownames(rawword)[-hereim] %>% 652 namedarows <- rownames(rawword)[-hereim] %>%
635 as.data.frame() 653 as.data.frame()
636 RAWWORD <- rawword[-hereim,] %>% 654 RAWWORD <- rawword[-hereim,] %>%
637 as.data.frame() %>% 655 as.data.frame() %>%
638 bind_cols(namedarows,.) 656 bind_cols(namedarows,.)
639 z <- 1 657 z <- 1
640 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE) 658 naroww <- as.data.frame(rep(0,dim(RAWWORD)[1]),stringsAsFactors = FALSE)
641 for(z in 1:dim(RAWWORD)[1]){ 659 for(z in 1:dim(RAWWORD)[1]){
642 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,]))) 660 naroww[z,1] <- as.integer(sum(is.na(RAWWORD[z,])))
643 z <- z + 1 661 z <- z + 1
644 } 662 }
645 663
646 colnames(naroww) <- "ROW_NAs" 664 colnames(naroww) <- "ROW_NAs"
647 RAWWORD <- bind_cols(RAWWORD,naroww) 665 RAWWORD <- bind_cols(RAWWORD,naroww)
648 666
649 667
650 roALZna <- t(ALZDAT) %>% 668 roALZna <- t(ALZDAT) %>%
651 rownames(.) %>% 669 rownames(.) %>%
652 as.data.frame(.) 670 as.data.frame(.)
653 colnames(roALZna) <- "ID_REF" 671 colnames(roALZna) <- "ID_REF"
654 672
655 RAWDAT <- t(ALZDAT) %>% 673 RAWDAT <- t(ALZDAT) %>%
656 as.data.frame(.) 674 as.data.frame(.)
657 colnames(RAWDAT) <- NULL 675 colnames(RAWDAT) <- NULL
658 rownames(RAWDAT) <- NULL 676 rownames(RAWDAT) <- NULL
659 677
660 RAWDAT2 <- RAWDAT %>% 678 RAWDAT2 <- RAWDAT %>%
661 cbind(roALZna,.) %>% 679 cbind(roALZna,.) %>%
662 dplyr::arrange(.,ID_REF) 680 dplyr::arrange(.,ID_REF)
663 681
664 ##Editing the file for R processing 682 ##Editing the file for R processing
665 RAWDATID <- RAWDAT2[,1] %>% 683 RAWDATID <- RAWDAT2[,1] %>%
666 as.matrix(.) 684 as.matrix(.)
667 685
668 RAWDATNUM <- RAWDAT2[,-1] %>% 686 RAWDATNUM <- RAWDAT2[,-1] %>%
669 mapply(.,FUN = as.numeric) %>% 687 mapply(.,FUN = as.numeric) %>%
670 t(.) 688 t(.)
671 689
672 ##Consolidating genes with the same name 690 ##Consolidating genes with the same name
673 ###create empty matrix of size equal to tabRDATID 691 ###create empty matrix of size equal to tabRDATID
674 tabRDATID <- table(RAWDATID) 692 tabRDATID <- table(RAWDATID)
675 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID)) 693 NuRDATN <- matrix(0, nrow = dim(RAWDATNUM)[1], ncol = length(tabRDATID))
676 j <- 1 694 j <- 1
677 for(j in 1:length(tabRDATID)){ 695 for(j in 1:length(tabRDATID)){
678 ##Putting the ones without duplicates in their new homes 696 ##Putting the ones without duplicates in their new homes
679 if(tabRDATID[j] == 1){ 697 if(tabRDATID[j] == 1){
680 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])] 698 NuRDATN[,j] <- RAWDATNUM[,which(RAWDATID==rownames(tabRDATID)[j])]
681 } 699 }
682 ##Averaging duplicates and putting them in their new homes 700 ##Averaging duplicates and putting them in their new homes
683 else if(tabRDATID[j] > 1){ 701 else if(tabRDATID[j] > 1){
684 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)
685 } 703 }
686 j <- j + 1 704 j <- j + 1
687 } 705 }
688 706
689 ##Scaling the Data 707 ##Scaling the Data
690 scrawdat <- NuRDATN%>% 708 scrawdat <- NuRDATN%>%
691 scale() 709 scale()
692 attr(scrawdat,"scaled:center") <- NULL 710 attr(scrawdat,"scaled:center") <- NULL
693 attr(scrawdat,"scaled:scale") <- NULL 711 attr(scrawdat,"scaled:scale") <- NULL
694 colnames(scrawdat) <- rownames(tabRDATID) 712 colnames(scrawdat) <- rownames(tabRDATID)
695 713
696 ##Discretized the Data 714 ##Discretized the Data
697 dialzdat <- scrawdat %>% 715 dialzdat <- scrawdat %>%
698 dndat(.) %>% 716 dndat(.) %>%
699 t()%>% 717 t()%>%
700 as.data.frame(.) 718 as.data.frame(.)
701 colnames(dialzdat) <- rownames(RAWDATNUM) 719 colnames(dialzdat) <- rownames(RAWDATNUM)
702 720
703 ##setting "ID_REF" as a new variable 721 ##setting "ID_REF" as a new variable
704 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1)) 722 geneNAM <- as.data.frame(as.matrix(rownames(dialzdat),ncol=1))
705 colnames(geneNAM) <- "ID_REF" 723 colnames(geneNAM) <- "ID_REF"
706 rownames(dialzdat) <- NULL 724 rownames(dialzdat) <- NULL
707 dialzdat <-bind_cols(geneNAM,dialzdat) 725 dialzdat <-bind_cols(geneNAM,dialzdat)
708 726
709 ##NAs in a column 727 ##NAs in a column
710 x <- 2 728 x <- 2
711 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)
712 nacol[1,1] = "COL_NAs" 730 nacol[1,1] = "COL_NAs"
713 for(x in 2:dim(dialzdat)[2]){ 731 for(x in 2:dim(dialzdat)[2]){
714 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x]))) 732 nacol[1,x] <- as.integer(sum(is.na(dialzdat[,x])))
715 x <- x + 1 733 x <- x + 1
716 } 734 }
717 colnames(nacol) <- colnames(dialzdat) 735 colnames(nacol) <- colnames(dialzdat)
718 dialzdat <- bind_rows(dialzdat,nacol) 736 dialzdat <- bind_rows(dialzdat,nacol)
719 737
720 ##NAs in a row 738 ##NAs in a row
721 y <- 1 739 y <- 1
722 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE) 740 narowd <- as.data.frame(rep(0,dim(dialzdat)[1]),stringsAsFactors = FALSE)
723 for(y in 1:dim(dialzdat)[1]){ 741 for(y in 1:dim(dialzdat)[1]){
724 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,]))) 742 narowd[y,1] <- as.integer(sum(is.na(dialzdat[y,])))
725 y <- y + 1 743 y <- y + 1
726 } 744 }
727 colnames(narowd) <- "ROW_NAs" 745 colnames(narowd) <- "ROW_NAs"
728 dialzdat <- bind_cols(dialzdat,narowd) 746 dialzdat <- bind_cols(dialzdat,narowd)
729 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam 747 colnames(dialzdat)[2:(dim(dialzdat)[2]-1)] <- subjnam
730 colnames(RAWWORD) <- colnames(dialzdat) 748 colnames(RAWWORD) <- colnames(dialzdat)
731 ##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
732 k <- 2 750 k <- 2
733 for(k in 2:dim(dialzdat)[2]-1){ 751 for(k in 2:dim(dialzdat)[2]-1){
734 dialzdat[,k] <- as.character(dialzdat[,k]) 752 dialzdat[,k] <- as.character(dialzdat[,k])
735 k <- k + 1 753 k <- k + 1
736 } 754 }
737 #The End the full data 755 #The End the full data
738 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat) 756 Dscrtalzdw <- bind_rows(RAWWORD,dialzdat)
739 757
740 #Produces Discrete file 758 #Produces Discrete file
741 nfnaex2 <- strsplit(alz,"[\\|/]") %>% 759 nfnaex2 <- strsplit(alz,"[\\|/]") %>%
742 .[[1]] %>% 760 .[[1]] %>%
743 .[length(.)] %>% 761 .[length(.)] %>%
744 gsub("\\D","",.) %>% 762 gsub("\\D","",.) %>%
745 c("GSE",.,"dscrt.txt") %>% 763 c("GSE",.,"dscrt.txt") %>%
746 paste(collapse = "") 764 paste(collapse = "")
747 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)
748 766
749 767
750 n <- n + 1 768 n <- n + 1
751 } 769 }
752 } 770 }
753 } 771 }
754 } 772 }