Efrain Gonzalez / Cleaning and Fixing Data with R

Download as
Browse Code ยป

Commit 679b62805671e900eb8f790c2866f091b576180d

Authored by Efrain Gonzalez
1 parent 6b8f81e7b3
Exists in master

completed code for path finding(TESTING)

Showing 1 changed file with 44 additions and 12 deletions   Show diff stats
1 #Efrain Gonzalez 1 #Efrain Gonzalez
2 #8/28/2017 2 #8/25/2017
3 #Code for Markov Blanket 3 #Code for Markov Blanket
4 4
5 5
6 #The required libraries 6 #The required libraries
7 library(pryr) 7 library(pryr)
8 library(MASS) 8 library(MASS)
9 library(dplyr) 9 library(dplyr)
10 library(tidyr) 10 library(tidyr)
11 library(readr) 11 library(readr)
12 library(stringr) 12 library(stringr)
13 13
14 14
15 #Have the user choose an original Dot file that they want to use 15 #Have the user choose an original Dot file that they want to use
16 DotFile <- file.choose() 16 DotFile <- file.choose()
17 TheDotP1 <- read_delim(DotFile,delim = "\t",col_names = FALSE) %>% 17 TheDotP1 <- read_delim(DotFile,delim = "\t",col_names = FALSE) %>%
18 dplyr::filter(!grepl("->|[{}]",X1)) %>% 18 dplyr::filter(!grepl("->|[{}]",X1)) %>%
19 dplyr::filter(!grepl("Banjo",X1)) %>% 19 dplyr::filter(!grepl("Banjo",X1)) %>%
20 dplyr::filter(!grepl("labeljust",X1)) 20 dplyr::filter(!grepl("labeljust",X1))
21 counterP1 <- 1 21 counterP1 <- 1
22 sizeDotP1 <- dim(TheDotP1)[1] 22 sizeDotP1 <- dim(TheDotP1)[1]
23 NewDotP1 <- matrix("0",ncol = 2, nrow = sizeDotP1) 23 NewDotP1 <- matrix("0",ncol = 2, nrow = sizeDotP1)
24 for(counterP1 in 1:sizeDotP1){ 24 for(counterP1 in 1:sizeDotP1){
25 coldataP1 <- str_trim(TheDotP1[counterP1,1]) %>% 25 coldataP1 <- str_trim(TheDotP1[counterP1,1]) %>%
26 as.character(.,stringsAsFactors = FALSE) 26 as.character(.,stringsAsFactors = FALSE)
27 if(grepl("Banjo|labeljust|>",coldataP1)==FALSE){ 27 if(grepl("Banjo|labeljust|>",coldataP1)==FALSE){
28 NumberP1 <- strsplit(coldataP1," ") %>% 28 NumberP1 <- strsplit(coldataP1," ") %>%
29 .[[1]]%>% 29 .[[1]]%>%
30 .[1] 30 .[1]
31 VarNameP1 <- strsplit(coldataP1," ") %>% 31 VarNameP1 <- strsplit(coldataP1," ") %>%
32 .[[1]] %>% 32 .[[1]] %>%
33 .[2] %>% 33 .[2] %>%
34 strsplit(.,"\"") %>% 34 strsplit(.,"\"") %>%
35 .[[1]] %>% 35 .[[1]] %>%
36 .[grep("^\\w|^\\d",.)] 36 .[grep("^\\w|^\\d",.)]
37 NewDotP1[counterP1,1] <- VarNameP1 37 NewDotP1[counterP1,1] <- VarNameP1
38 NewDotP1[counterP1,2] <- NumberP1 38 NewDotP1[counterP1,2] <- NumberP1
39 } 39 }
40 if(grepl("->",coldataP1) == TRUE){ 40 if(grepl("->",coldataP1) == TRUE){
41 break 41 break
42 } 42 }
43 } 43 }
44 44
45 45
46 TheDotP2 <- read_delim(DotFile,delim = "\t",col_names = FALSE) %>% 46 TheDotP2 <- read_delim(DotFile,delim = "\t",col_names = FALSE) %>%
47 dplyr::filter(grepl("->",X1)) 47 dplyr::filter(grepl("->",X1))
48 counterP2 <- 1 48 counterP2 <- 1
49 sizeDotP2 <- dim(TheDotP2)[1] 49 sizeDotP2 <- dim(TheDotP2)[1]
50 NewDotP2 <- matrix("0",ncol = 2, nrow = sizeDotP2) 50 NewDotP2 <- matrix("0",ncol = 2, nrow = sizeDotP2)
51 for(counterP2 in 1:sizeDotP2){ 51 for(counterP2 in 1:sizeDotP2){
52 coldataP2 <- str_trim(TheDotP2[counterP2,1]) %>% 52 coldataP2 <- str_trim(TheDotP2[counterP2,1]) %>%
53 as.character(.,stringsAsFactors = FALSE) 53 as.character(.,stringsAsFactors = FALSE)
54 ParentNumP2 <- strsplit(coldataP2,"->") %>% 54 ParentNumP2 <- strsplit(coldataP2,"->") %>%
55 .[[1]]%>% 55 .[[1]]%>%
56 .[1] 56 .[1]
57 ChildNumP2 <- strsplit(coldataP2,"->") %>% 57 ChildNumP2 <- strsplit(coldataP2,"->") %>%
58 .[[1]] %>% 58 .[[1]] %>%
59 .[2] %>% 59 .[2] %>%
60 strsplit(.,";") %>% 60 strsplit(.,";") %>%
61 .[[1]] %>% 61 .[[1]] %>%
62 .[1] 62 .[1]
63 NewDotP2[counterP2,1] <- ParentNumP2 63 NewDotP2[counterP2,1] <- ParentNumP2
64 NewDotP2[counterP2,2] <- ChildNumP2 64 NewDotP2[counterP2,2] <- ChildNumP2
65 } 65 }
66 66
67 colnames(NewDotP2) <- c("Parents","Children") 67 colnames(NewDotP2) <- c("Parents","Children")
68 68
69 #Matching numbers to variable names 69 #Matching numbers to variable names
70 NewDotP2_2 <- NewDotP2 70 NewDotP2_2 <- NewDotP2
71 for(i in 1:sizeDotP1){ 71 for(i in 1:sizeDotP1){
72 #Where is the variable located within NewDotP2 (column one only)? 72 #Where is the variable located within NewDotP2 (column one only)?
73 chngreq <- grep(paste0("^",NewDotP1[i,2],"$"),NewDotP2_2[,1]) 73 chngreq <- grep(paste0("^",NewDotP1[i,2],"$"),NewDotP2_2[,1])
74 if(is.na(sum(chngreq)) == FALSE){ 74 if(is.na(sum(chngreq)) == FALSE){
75 if(sum(chngreq) > 0){ 75 if(sum(chngreq) > 0){
76 NewDotP2_2[chngreq,1] <- gsub(paste0("^",NewDotP1[i,2],"$"),NewDotP1[i,1],NewDotP2_2[chngreq,1]) 76 NewDotP2_2[chngreq,1] <- gsub(paste0("^",NewDotP1[i,2],"$"),NewDotP1[i,1],NewDotP2_2[chngreq,1])
77 } 77 }
78 } 78 }
79 #i <- i + 1 79 #i <- i + 1
80 } 80 }
81 NewDotP2_2 81 NewDotP2_2
82 for(j in 1:sizeDotP1){ 82 for(j in 1:sizeDotP1){
83 #Where is the variable located within NewDotP2 (column two only)? 83 #Where is the variable located within NewDotP2 (column two only)?
84 chngreq <- grep(paste0("^",NewDotP1[j,2],"$"),NewDotP2_2[,2]) 84 chngreq <- grep(paste0("^",NewDotP1[j,2],"$"),NewDotP2_2[,2])
85 if(is.na(sum(chngreq)) == FALSE){ 85 if(is.na(sum(chngreq)) == FALSE){
86 if(sum(chngreq) > 0){ 86 if(sum(chngreq) > 0){
87 NewDotP2_2[chngreq,2] <- gsub(paste0("^",NewDotP1[j,2],"$"),NewDotP1[j,1],NewDotP2_2[chngreq,2]) 87 NewDotP2_2[chngreq,2] <- gsub(paste0("^",NewDotP1[j,2],"$"),NewDotP1[j,1],NewDotP2_2[chngreq,2])
88 } 88 }
89 } 89 }
90 #j <- j + 1 90 #j <- j + 1
91 } 91 }
92 92
93 lrgMarkov <- dim(NewDotP2_2)[1] 93 lrgMarkov <- dim(NewDotP2_2)[1]
94 Blanky <- function(MarkovDegree = 20, VariableStartName = "Alzheimer", VariableEndName = "GRIN2A"){ 94 Blanky <- function(MarkovDegree = 20, VariableStartName = "Alzheimer", VariableEndName = "GRIN2A"){
95 #Finding the Parents and Children 95 #Finding the Parents and Children
96 d <- 1 96 d <- 1
97 AllNamList1 <- vector("list",length = 3) 97 AllNamList1 <- vector("list",length = 3)
98 #AllNamList <- vector("list", length = 3) 98 #AllNamList <- vector("list", length = 3)
99 ##Finding the Parents of the Parents 99 ##Finding the Parents of the Parents
100 ##A list of lists 100 ##A list of lists
101 ##outer set by the degree of the Markov blanket 101 ##outer set by the degree of the Markov blanket
102 AllVarList <- vector("list",length = 3) 102 AllVarList <- vector("list",length = 3)
103 varfound <- 0 103 varfound <- 0
104 for(d in 1:MarkovDegree){ 104 for(d in 1:MarkovDegree){
105 colnames(NewDotP2_2) <- NULL 105 colnames(NewDotP2_2) <- NULL
106 ##Which variable are you looking for? 106 ##Which variable are you looking for?
107 ##This is the VariableEndName 107 ##This is the VariableEndName
108 if(d == 1){ 108 if(d == 1){
109 ##Finding the Parents for the variable 109 ##Finding the Parents for the variable
110 LocPofVar <- grep(paste0("^",VariableStartName,"$"),NewDotP2_2[,2]) 110 LocPofVar <- grep(paste0("^",VariableStartName,"$"),NewDotP2_2[,2])
111 PofVar <- NewDotP2_2[LocPofVar,1] 111 PofVar <- NewDotP2_2[LocPofVar,1]
112 AllNamList1[[1]] <- PofVar 112 AllNamList1[[1]] <- PofVar
113 AllNamList1[[1]] <- AllNamList1[[1]][!duplicated(AllNamList1[[1]])] 113 AllNamList1[[1]] <- AllNamList1[[1]][!duplicated(AllNamList1[[1]])]
114 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[1]])) 114 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[1]]))
115 115
116 ##Finding the Children for the variable 116 ##Finding the Children for the variable
117 LocCofVar <- grep(paste0("^",VariableStartName,"$"),NewDotP2_2[,1]) 117 LocCofVar <- grep(paste0("^",VariableStartName,"$"),NewDotP2_2[,1])
118 CofVar <- NewDotP2_2[LocCofVar,2] 118 CofVar <- NewDotP2_2[LocCofVar,2]
119 AllNamList1[[2]] <- CofVar 119 AllNamList1[[2]] <- CofVar
120 AllNamList1[[2]] <- AllNamList1[[2]][!duplicated(AllNamList1[[2]])] 120 AllNamList1[[2]] <- AllNamList1[[2]][!duplicated(AllNamList1[[2]])]
121 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[2]])) 121 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[2]]))
122 122
123 ##Finding the Co-Parents of the Children for the variable 123 ##Finding the Co-Parents of the Children for the variable
124 NumofChild <- length(CofVar) 124 NumofChild <- length(CofVar)
125 if(NumofChild > 0){ 125 if(NumofChild > 0){
126 ##Creating a list of the Co-Parents for each of the children 126 ##Creating a list of the Co-Parents for each of the children
127 ##list size is based on the amount of Children 127 ##list size is based on the amount of Children
128 COPlist <- vector("character", length = 0) 128 COPlist <- vector("character", length = 0)
129 nc <- 1 129 nc <- 1
130 for(nc in 1:NumofChild){ 130 for(nc in 1:NumofChild){
131 LocCOPofVar <- grep(paste0("^",CofVar[nc],"$"),NewDotP2_2[,2]) 131 LocCOPofVar <- grep(paste0("^",CofVar[nc],"$"),NewDotP2_2[,2])
132 COPofVar <- NewDotP2_2[LocCOPofVar,1] 132 COPofVar <- NewDotP2_2[LocCOPofVar,1]
133 if(sum(grepl(VariableStartName,COPofVar)) >= 1){ 133 if(sum(grepl(VariableStartName,COPofVar)) >= 1){
134 #positions of variable start name within the vector of co parents 134 #positions of variable start name within the vector of co parents
135 posoforig <- grep(paste0("^",VariableStartName,"$"),COPofVar) 135 posoforig <- grep(paste0("^",VariableStartName,"$"),COPofVar)
136 COPofVar <- COPofVar[-posoforig] 136 COPofVar <- COPofVar[-posoforig]
137 COPlist <- append(COPlist,COPofVar) 137 COPlist <- append(COPlist,COPofVar)
138 } else{ 138 } else{
139 #COPlist[[nc]] <- COPofVar[COPofVar!=VariableStartName] 139 #COPlist[[nc]] <- COPofVar[COPofVar!=VariableStartName]
140 COPlist <- append(COPlist,COPofVar) 140 COPlist <- append(COPlist,COPofVar)
141 } 141 }
142 #nc <- nc + 1 142 #nc <- nc + 1
143 } 143 }
144 } else { 144 } else {
145 ##Making COPlist empty 145 ##Making COPlist empty
146 COPlist <- vector("character",length = 0) 146 COPlist <- vector("character",length = 0)
147 } 147 }
148 AllNamList1[[3]] <- COPlist 148 AllNamList1[[3]] <- COPlist
149 AllNamList1[[3]] <- AllNamList1[[3]][!duplicated(AllNamList1[[3]])] 149 AllNamList1[[3]] <- AllNamList1[[3]][!duplicated(AllNamList1[[3]])]
150 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[3]])) 150 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllNamList1[[3]]))
151 151
152 AllVarList[[1]] <- AllNamList1 152 AllVarList[[1]] <- AllNamList1
153 153
154 } else if(d > 1){ 154 } else if(d > 1){
155 ##inner set by the length of the previous AllVarlist we are working on 155 ##inner set by the length of the previous AllVarlist we are working on
156 lPreVList <- length(AllVarList[[d-1]]) 156 lPreVList <- length(AllVarList[[d-1]])
157 ef <- 1 157 ef <- 1
158 ##PCCP will eventually equal the total size that we expect for the iteration (#d) 158 ##PCCP will eventually equal the total size that we expect for the iteration (#d)
159 ## which is just lPreVList * 3 159 ## which is just lPreVList * 3
160 PCCP <- 1 160 PCCP <- 1
161 newsize <- (lPreVList * 3) 161 newsize <- (lPreVList * 3)
162 if(d > 3){ 162 if(d > 3){
163 AllVarList[[d]] <- vector("list") 163 AllVarList[[d]] <- vector("list")
164 } 164 }
165 for(ef in 1:lPreVList){ 165 for(ef in 1:lPreVList){
166 ##Finding the Parents 166 ##Finding the Parents
167 NumofVars <- length(AllVarList[[d-1]][[ef]]) 167 NumofVars <- length(AllVarList[[d-1]][[ef]])
168 if(NumofVars > 0){ 168 if(NumofVars > 0){
169 ##Creating a list of the Parents for each of the Variables 169 ##Creating a list of the Parents for each of the Variables
170 ##list size is based on the amount of Previous Variables 170 ##list size is based on the amount of Previous Variables
171 PofVlist <- vector("character", length = 0) 171 PofVlist <- vector("character", length = 0)
172 np <- 1 172 np <- 1
173 for(np in 1:NumofVars){ 173 for(np in 1:NumofVars){
174 LocPofVar <- grep(paste0("^",AllVarList[[d-1]][[ef]][np],"$"),NewDotP2_2[,2]) 174 LocPofVar <- grep(paste0("^",AllVarList[[d-1]][[ef]][np],"$"),NewDotP2_2[,2])
175 PofVar <- NewDotP2_2[LocPofVar,1] 175 PofVar <- NewDotP2_2[LocPofVar,1]
176 PofVlist <- append(PofVlist,PofVar) 176 PofVlist <- append(PofVlist,PofVar)
177 #np <- np + 1 177 #np <- np + 1
178 } 178 }
179 } else { 179 } else {
180 ##Making COPlist empty 180 ##Making COPlist empty
181 PofVlist <- vector("character",length = 0) 181 PofVlist <- vector("character",length = 0)
182 } 182 }
183 AllVarList[[d]][[PCCP]] <- PofVlist 183 AllVarList[[d]][[PCCP]] <- PofVlist
184 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])] 184 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])]
185 ##Have you found the VariableEndName? 185 ##Have you found the VariableEndName?
186 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]])) 186 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]]))
187 PCCP <- PCCP + 1 187 PCCP <- PCCP + 1
188 188
189 ##Finding the Children 189 ##Finding the Children
190 if(NumofVars > 0){ 190 if(NumofVars > 0){
191 ##Creating a list of the Children for each of the Previous Parents 191 ##Creating a list of the Children for each of the Previous Parents
192 ##list size is based on the amount of Previous Parents 192 ##list size is based on the amount of Previous Parents
193 CofVlist <- vector("character", length = 0) 193 CofVlist <- vector("character", length = 0)
194 np <- 1 194 np <- 1
195 for(np in 1:NumofVars){ 195 for(np in 1:NumofVars){
196 LocCofVar <- grep(paste0("^",AllVarList[[d-1]][[ef]][np],"$"),NewDotP2_2[,1]) 196 LocCofVar <- grep(paste0("^",AllVarList[[d-1]][[ef]][np],"$"),NewDotP2_2[,1])
197 CofVar <- NewDotP2_2[LocCofVar,2] 197 CofVar <- NewDotP2_2[LocCofVar,2]
198 #if(sum(grepl(VariableStartName,CofVar)) >= 1){ 198 #if(sum(grepl(VariableStartName,CofVar)) >= 1){
199 # #positions of variable start name within the vector of co parents 199 # #positions of variable start name within the vector of co parents
200 # posoforig <- grep(VariableStartName,COPofVar) 200 # posoforig <- grep(VariableStartName,COPofVar)
201 # COPofVar <- COPofVar[-posoforig] 201 # COPofVar <- COPofVar[-posoforig]
202 # COPlist <- append(COPlist,COPofVar) 202 # COPlist <- append(COPlist,COPofVar)
203 #} else{ 203 #} else{
204 CofVlist <- append(CofVlist,CofVar) 204 CofVlist <- append(CofVlist,CofVar)
205 #} 205 #}
206 #np <- np + 1 206 #np <- np + 1
207 } 207 }
208 } else { 208 } else {
209 ##Making CofPlist empty 209 ##Making CofPlist empty
210 CofVlist <- vector("character",length = 0) 210 CofVlist <- vector("character",length = 0)
211 } 211 }
212 AllVarList[[d]][[PCCP]] <- CofVlist 212 AllVarList[[d]][[PCCP]] <- CofVlist
213 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])] 213 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])]
214 ##Have you found the VariableEndName yet? 214 ##Have you found the VariableEndName yet?
215 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]])) 215 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]]))
216 PCCP <- PCCP + 1 216 PCCP <- PCCP + 1
217 217
218 ##Finding the Co-Parents 218 ##Finding the Co-Parents
219 NumofCVars <- length(CofVlist) 219 NumofCVars <- length(CofVlist)
220 if(NumofCVars > 0){ 220 if(NumofCVars > 0){
221 ncp <- 1 221 ncp <- 1
222 CPofClist <- vector("character",length = 0) 222 CPofClist <- vector("character",length = 0)
223 for(ncp in 1:NumofCVars){ 223 for(ncp in 1:NumofCVars){
224 LocCPofCVar <- grep(paste0("^",CofVlist[ncp],"$"),NewDotP2_2[,2]) 224 LocCPofCVar <- grep(paste0("^",CofVlist[ncp],"$"),NewDotP2_2[,2])
225 CPofCVar <- NewDotP2_2[LocCPofCVar,1] 225 CPofCVar <- NewDotP2_2[LocCPofCVar,1]
226 #if(sum(grepl(,CPofCVar)) >= 1){ 226 #if(sum(grepl(,CPofCVar)) >= 1){
227 # #positions of variable start name within the vector of co parents 227 # #positions of variable start name within the vector of co parents
228 # posoforig <- grep(VariableStartName,COPofVar) 228 # posoforig <- grep(VariableStartName,COPofVar)
229 # COPofVar <- COPofVar[-posoforig] 229 # COPofVar <- COPofVar[-posoforig]
230 # COPlist <- append(COPlist,COPofVar) 230 # COPlist <- append(COPlist,COPofVar)
231 #} else{ 231 #} else{
232 CPofClist <- append(CPofClist,CPofCVar) 232 CPofClist <- append(CPofClist,CPofCVar)
233 #} 233 #}
234 #ncp <- ncp + 1 234 #ncp <- ncp + 1
235 } 235 }
236 236
237 } else { 237 } else {
238 ##Making COPlist empty 238 ##Making COPlist empty
239 CPofClist <- vector("character",length = 0) 239 CPofClist <- vector("character",length = 0)
240 } 240 }
241 AllVarList[[d]][[PCCP]] <- CPofClist 241 AllVarList[[d]][[PCCP]] <- CPofClist
242 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])] 242 AllVarList[[d]][[PCCP]] <- AllVarList[[d]][[PCCP]][!duplicated(AllVarList[[d]][[PCCP]])]
243 ##Have you found VariableEndName now? 243 ##Have you found VariableEndName now?
244 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]])) 244 varfound <- varfound + sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[PCCP]]))
245 PCCP <- PCCP + 1 245 PCCP <- PCCP + 1
246 #ef <- ef + 1 246 #ef <- ef + 1
247 } 247 }
248 } 248 }
249 249
250 ##Stop if you have found the VariableEndName value 250 ##Stop if you have found the VariableEndName value
251 if(varfound > 0){ 251 if(varfound > 0){
252 break 252 break
253 } 253 }
254 #d <- d + 1 254 #d <- d + 1
255 } 255 }
256 ##The Markov Degree is that found below 256 ##The Markov Degree is that found below
257 d 257 d
258 } 258 }
259 259
260 260
261 261
262 #############Methods for finding paths############# 262 #############Methods for finding paths#############
263 ###Beyond Here Everything is still experimental### 263 ###Beyond Here Everything is still experimental###
264 #Find the path to the variable in question 264 #Find the path to the variable in question
265 RshipList <- vector("list",length = d) 265 RshipList <- vector("list",length = d)
266 l = d 266 l = d
267 for(l in d:1){ 267 for(l in d:1){
268 if(l == d){ 268 if(l == d){
269 RshipList[[l]] <- VariableEndName 269 RshipList[[l]] <- VariableEndName
270 for(ship in 1:length(AllVarList[[d]])){ 270 for(ship in 1:length(AllVarList[[d]])){
271 if(sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[ship]])) > 0){ 271 if(sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[ship]])) > 0){
272 break 272 break
273 } 273 }
274 } 274 }
275 275
276 } else { 276 } else {
277 modship <- ship %% 3 277 modship <- ship %% 3
278 intship <- as.integer(ship/3) 278 intship <- as.integer(ship/3)
279 if(modship > 0){ 279 if(modship > 0){
280 intship <- intship + 1 280 intship <- intship + 1
281 } 281 }
282 RshipList[[l]] <- AllVarList[[l]][[intship]] 282 RshipList[[l]] <- AllVarList[[l]][[intship]]
283 ship <- intship 283 ship <- intship
284 } 284 }
285 #l <- l - 1 285 #l <- l - 1
286 } 286 }
287 #Full actual path 287 #Full actual path
288 k = d 288 k = d
289 prevar <- VariableEndName 289 prevar <- VariableEndName
290 Pathlisting <- vector("list",length = d) 290 Pathlisting <- vector("list",length = d)
291 Pathlisting[[d]] <- prevar 291 Pathlisting[[d]] <- prevar
292 for(k in d:1){ 292 for(k in d:2){
293 if(k == d){ 293 if(k == d){
294 for(ship in 1:length(AllVarList[[d]])){ 294 for(ship in 1:length(AllVarList[[d]])){
295 if(sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[ship]])) > 0){ 295 if(sum(grepl(paste0("^",VariableEndName,"$"),AllVarList[[d]][[ship]])) > 0){
296 break 296 break
297 } 297 }
298 } 298 }
299 299
300 } 300 }
301 modship <- ship %% 3 301 modship <- ship %% 3
302 intship <- as.integer(ship/3) 302 intship <- as.integer(ship/3)
303 if(modship > 0){ 303 if(modship > 0){
304 intship <- intship + 1 304 intship <- intship + 1
305 } 305 }
306 if(modship == 0){ 306 if(modship == 0){
307 ##When modship == 0 then we are refering to a CoParent 307 ##When modship == 0 then we are refering to a CoParent
308 ##Gives how many children the full set of CoParents has 308 ##Gives how many children the full set of CoParents has
309 lenCoP <- length(AllVarList[[k]][[(ship - 1)]]) 309 lenCoP <- length(AllVarList[[k]][[(ship - 1)]])
310 ##variables that are actually children of the coparent 310 ##variables that are actually children of the coparent
311 actvar <- vector("character",length = 0) 311 actvar <- vector("character",length = 0)
312 ##Parents of actvar 312 ##Parents of actvar
313 Pactvar <- vector("character",length =0) 313 Pactvar <- vector("character",length =0)
314 ##variables that could have led to the previous set of variables 314 ##variables that could have led to the previous set of variables
315 Wanvar <- vector("character",length = 0) 315 Wanvar <- vector("character",length = 0)
316 m <- 1 316 m <- 1
317 for(m in 1:lenCoP){ 317 for(m in 1:lenCoP){
318 LCPoVar <- grep(paste0("^",AllVarList[[k]][[(ship - 1)]][m],"$"),NewDotP2_2[,2]) 318 LCPoVar <- grep(paste0("^",AllVarList[[k]][[(ship - 1)]][m],"$"),NewDotP2_2[,2])
319 CPoVar <- NewDotP2_2[LCPoVar,1] 319 CPoVar <- NewDotP2_2[LCPoVar,1]
320 lenprevar <- length(prevar) 320 lenprevar <- length(prevar)
321 y <- 1 321 y <- 1
322 for(y in 1:lenprevar){ 322 for(y in 1:lenprevar){
323 323
324 if(sum(grepl(prevar[y],CPoVar)) >= 1){ 324 if(sum(grepl(prevar[y],CPoVar)) >= 1){
325 actvar <- append(actvar,AllVarList[[k]][[(ship - 1)]][m]) 325 actvar <- append(actvar,AllVarList[[k]][[(ship - 1)]][m])
326 Pactvar <- append(Pactvar,CPoVar) 326 Pactvar <- append(Pactvar,CPoVar)
327 } 327 }
328 #y <- y + 1 328 #y <- y + 1
329 } 329 }
330 #m <- m + 1 330 #m <- m + 1
331 331
332 } 332 }
333 Pactvar <- Pactvar[!duplicated(Pactvar)] 333 Pactvar <- Pactvar[!duplicated(Pactvar)]
334 ##Searching to see if any of the parents are in the previous degree 334 ##Searching to see if any of the parents are in the previous degree
335 ##The right children will have both the current variable and a previous degrees variable as parents 335 ##The right children will have both the current variable and a previous degrees variable as parents
336 o <- 1 336 o <- 1
337 for(o in 1:length(Pactvar)){ 337 for(o in 1:length(Pactvar)){
338 if(sum(grepl(Pactvar[o],AllVarList[[(k - 1)]][[intship]])) >= 1){ 338 if(sum(grepl(Pactvar[o],AllVarList[[(k - 1)]][[intship]])) >= 1){
339 Wanvar <- append(Wanvar,Pactvar[o]) 339 Wanvar <- append(Wanvar,Pactvar[o])
340 } 340 }
341 #o <- o + 1 341 #o <- o + 1
342 } 342 }
343 Wanvar <- Wanvar[!duplicated(Wanvar)] 343 Wanvar <- Wanvar[!duplicated(Wanvar)]
344 prevar <- Wanvar 344 prevar <- Wanvar
345 345
346 } else if(modship == 1){ 346 } else if(modship == 1){
347 #When modship == 1 then we are referring to a Parent 347 #When modship == 1 then we are referring to a Parent
348 LCofVar <- grep(paste0("^",prevar,"$"),NewDotP2_2[,1]) 348 lengprevar <- length(prevar)
349 CoVar <- NewDotP2_2[LCofVar,1] 349 t <- 1
350 lenC <- length(CoVar) 350 ##All children of these parents
351 for(o in 1:lenC){ 351 TCoVar <- vector("character",length = 0)
352 if(grepl(CoVar[o],AllVarList[[k]][[intship]]) == TRUE){ 352 ##variables that could have led to these parents
353 353 leadVar <- vector("character",length = 0)
354 354 for(t in 1: lengprevar){
355 LCofVar <- grep(paste0("^",prevar[t],"$"),NewDotP2_2[,1])
356 CoVar <- NewDotP2_2[LCofVar,2]
357 TCoVar <- append(TCoVar,CoVar)
358 }
359 TCoVar <- TCoVar[!duplicated(TCoVar)]
360 s <- 1
361 ##Searching for the children that are also in the previous degree
362 for(s in 1:length(TCoVar)){
363 if(sum(grepl(TCoVar[s],AllVarList[[(k - 1)]][[intship]])) >= 1){
364 leadVar <- append(leadVar,TCoVar[s])
355 } 365 }
356
357 } 366 }
358 367 leadVar <- leadVar[!duplicated(leadVar)]
368 prevar <- leadVar
359 369
360 } else if(modship == 2){ 370 } else if(modship == 2){
361 #When modship == 0 then we are referring to a Child 371 #When modship == 2 then we are referring to a Child
372 lenprevar <- length(prevar)
373 x <- 1
374 ##All parents of these children
375 TPoVar <- vector("character",length = 0)
376 ##variables that could have led to these children based on the
377 drctVar <- vector("character",length = 0)
378 for(x in 1:lenprevar){
379 LPoVar <- grep(paste0("^",prevar[x],"$"),NewDotP2_2[,2])
380 PoVar <- NewDotP2_2[LPoVar,1]
381 TPoVar <- append(TPoVar,PoVar)
382 }
383 TPoVar <- TPoVar[!duplicated(TPoVar)]
384 y <- 1
385 ##Searching for the parents that are also in the previous degree
386 for(y in 1:length(TPoVar)){
387 if(sum(grepl(TPoVar[y],AllVarList[[(k - 1)]][[intship]])) >= 1){
388 drctVar <- append(drctVar,TPoVar[y])
389 }
390 }
391 drctVar <- drctVar[!duplicated(drctVar)]
392 prevar <- drctVar
393
362 394
363 } 395 }
364 396
365 ship <- intship 397 ship <- intship
366 398
367 Pathlisting[[(k - 1)]] <- prevar 399 Pathlisting[[(k - 1)]] <- prevar
368 } 400 }