Mercurial > repos > jfb > commonality_finder

comparison C and D finder/commonality-working feb 21 20.R @ 4:830e8dc3acac draft default tip

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
Uploaded
author jfb
date Tue, 14 Jul 2020 19:51:12 -0400
parents 6eacb089b655
children
comparison
equal deleted inserted replaced
3:6eacb089b655 4:830e8dc3acac
1 FirstSubstrateSet<- read.csv("input1.csv", stringsAsFactors=FALSE, header = FALSE)
2 Firstsubbackfreq<- read.csv("input2.csv", header=FALSE, stringsAsFactors=FALSE)
3 SubstrateHeader<-FirstSubstrateSet[1,]
4 FirstSubstrateSet<- FirstSubstrateSet[2:nrow(FirstSubstrateSet),]
5 if(nrow(Firstsubbackfreq[1,]>35)){
6 if(grepl(pattern = "Properties", x=Firstsubbackfreq[1,22])){
7 Firstsubbackfreq<-t(Firstsubbackfreq)
8 }
9 }
10
11
12 SecondSubstrateSet<- read.csv("input3.csv", stringsAsFactors=FALSE, header = FALSE)
13 Secondsubbackfreq<- read.csv("input4.csv", header=FALSE, stringsAsFactors=FALSE)
14 SecondSubstrateSet<- SecondSubstrateSet[2:nrow(SecondSubstrateSet),]
15 if(nrow(Secondsubbackfreq[1,]>35)){
16 if(grepl(pattern = "Properties", x=Secondsubbackfreq[1,22])){
17 Secondsubbackfreq<-t(Secondsubbackfreq)
18 }
19 }
20
21
22 ThirdSubstrateSet<- read.csv("input5.csv", stringsAsFactors=FALSE, header = FALSE)
23 Thirdsubbackfreq<- read.csv("input6.csv", header=FALSE, stringsAsFactors=FALSE)
24 ThirdSubstrateSet<- ThirdSubstrateSet[2:nrow(ThirdSubstrateSet),]
25 if(nrow(Thirdsubbackfreq[1,]>35)){
26 if(grepl(pattern = "Properties", x=Thirdsubbackfreq[1,22])){
27 Thirdsubbackfreq<-t(Thirdsubbackfreq)
28 }
29 }
30
31
32
33
34 #ff you want ONLY FULL MOTIFS, put "YES" here, please use all caps
35 FullMotifsOnly_questionmark<-"NO"
36 #If you want ONLY TRUNCATED MOTIFS, put "YES" here, please use all caps
37 TruncatedMotifsOnly_questionmark<-"NO"
38 #if you want to find the overlap, put a "YES" here (all caps), if you want to find the non-overlap, put "NO" (all caps)
39 Are_You_Looking_For_Commonality<-"YES"
40
41
42 #then put the names of your output files here
43 Shared_motifs_table<-"sharedmotifs.csv"
44 Shared_subbackfreq_table<-"sharedSBF.csv"
45
46 # Shared_motifs_table<-"Shared motifs 7-27-17.csv"
47 # Shared_subbackfreq_table<-"SubstrateBackgrounFrequency-for-shared-motifs 4 7-27-17.csv"
48
49 First_unshared_motifs_table<-"R1 substrates.csv"
50 First_unshared_subbackfreq<-"R1 SBF.csv"
51
52 Second_unshared_motifs_table<-"R2 subs.csv"
53 Second_unshared_subbackfreq<-"R2 SBf.csv"
54
55 Third_unshared_motifs_table<-"R3 subs.csv"
56 Third_unshared_subbackfreq<-"R3 SBF.csv"
57
58 #final note, this code is going to be unworkable if you want to make a Venn diagram of more than 3 circles. I think I'll poke around
59 #other languages to see if any of them can do it.
60 ####################################################################################################################################
61
62
63
64
65
66 FirstxY<-rep("xY",times=nrow(FirstSubstrateSet))
67 FirstSubstrateSet[,11]<-FirstxY
68
69 SecondxY<-rep("xY",times=nrow(SecondSubstrateSet))
70 SecondSubstrateSet[,11]<-SecondxY
71
72 ThirdxY<-rep("xY",times=nrow(ThirdSubstrateSet))
73 ThirdSubstrateSet[,11]<-ThirdxY
74
75
76
77
78
79
80
81
82
83
84
85 ####################################################################################################################################
86 ####################################################################################################################################
87 # better version of this code written in C: what happens when two kinases share a motif, but they found that motif in two
88 # separate proteins thus two separate accession numbers?
89 # It should actually output the shared motif and BOTH accession numbers. Right now it does not, it only maps out the second
90 # accession number. So that needs to be fixed BUT you need to keep the commonality between a motif and its accession number
91 ####################################################################################################################################
92 ####################################################################################################################################
93 ####################################################################################################################################
94 ####################################################################################################################################
95
96 #Create the motif sets, deciding wether or not you're looking for truncated or full here
97 #full only
98
99 FTLwtmotifs=matrix(,nrow = nrow(FirstSubstrateSet),ncol=1)
100 FTLwtAccessionNumbers=matrix(data = Firstsubbackfreq[1,],ncol=1)
101
102 for (i in 1:nrow(FirstSubstrateSet)){
103 FTLwtletters<-FirstSubstrateSet[i,4:18]
104 FTLwtletters<-FTLwtletters[FTLwtletters !="XXXXX"]
105 FTLwtletters<-paste(FTLwtletters, sep="", collapse="")
106 leftspaces<-c()
107 rightspaces<-c()
108
109 YYYmotif <- unlist(strsplit(FTLwtletters, split = ""))
110 YYYposition <- match(x = "x", table = YYYmotif)
111 #position itself tells me how much is to the left of that X by what it's number is. x at position 4 tells me that there are
112 #just 3 letters to the left of x
113
114 YYYLettersToTheLeft <- YYYposition - 1
115 #how many letters to the right SHOULD just be length(motif)-position-1 if it's 5 long and x is at 3 then Y is at 4 and there is
116 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
117 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
118 #then sanity check, we're currently looking only at +/-4, but this spot allows for up to +/- 7 as well, just depends on what the
119 #variable the user puts in is
120
121
122 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
123 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
124 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
125 #add blank spaces if the motif has less than 4 letters to the left/right
126 motif<-c(leftspaces,YYYmotif,rightspaces)
127 #save that motif, which is the Y and +/- 4 amino acids, including truncation
128 motif<-motif[!motif %in% "x"]
129 motif<-paste(motif, sep="", collapse="")
130 FTLwtletters<-motif
131 FTLwtmotifs[i,1]<-FTLwtletters
132 # FTLwtAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
133 }
134
135 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
136 motif<-YYYmotif
137 #add blank spaces if the motif has less than 4 letters to the left/right
138 motif<-c(leftspaces,YYYmotif,rightspaces)
139 #save that motif, which is the Y and +/- 4 amino acids, including truncation
140 motif<-motif[!motif %in% "x"]
141 motif<-paste(motif, sep="", collapse="")
142 FTLwtletters<-motif
143 FTLwtmotifs[i,1]<-FTLwtletters
144 # FTLwtAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
145
146
147 }
148
149 }
150
151 D835Ymotifs=matrix(,nrow = nrow(SecondSubstrateSet),ncol=1)
152 D835YAccessionNumbers<-matrix(data = Secondsubbackfreq[1,],ncol = 1)
153
154 for (i in 1:nrow(SecondSubstrateSet)){
155 D835letters<-SecondSubstrateSet[i,4:18]
156 D835letters<-D835letters[D835letters !="XXXXX"]
157 D835letters<-paste(D835letters, sep="", collapse="")
158 leftspaces<-c()
159 rightspaces<-c()
160
161 YYYmotif <- unlist(strsplit(D835letters, split = ""))
162 YYYposition <- match(x = "x", table = YYYmotif)
163 #position itself tells me how much is to the left of that X by what it's number is. x at position 4 tells me that there are
164 #just 3 letters to the left of x
165
166 YYYLettersToTheLeft <- YYYposition - 1
167 #how many letters to the right SHOULD just be length(motif)-position-1 if it's 5 long and x is at 3 then Y is at 4 and there is
168 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
169 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
170 #then sanity check, we're currently looking only at +/-4, but this spot allows for up to +/- 7 as well, just depends on what the
171 #variable the user puts in is
172 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
173 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
174 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
175 #add blank spaces if the motif has less than 4 letters to the left/right
176 motif<-c(leftspaces,YYYmotif,rightspaces)
177 #save that motif, which is the Y and +/- 4 amino acids, including truncation
178 motif<-motif[!motif %in% "x"]
179 motif<-paste(motif, sep="", collapse="")
180 D835letters<-motif
181 D835Ymotifs[i,1]<-D835letters
182 # D835YAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
183 }
184
185 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
186 motif<-YYYmotif
187 #add blank spaces if the motif has less than 4 letters to the left/right
188 motif<-c(leftspaces,YYYmotif,rightspaces)
189 #save that motif, which is the Y and +/- 4 amino acids, including truncation
190 motif<-motif[!motif %in% "x"]
191 motif<-paste(motif, sep="", collapse="")
192 D835letters<-motif
193 D835Ymotifs[i,1]<-D835letters
194 # D835YAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
195 }
196 }
197
198
199 ITDmotifs=matrix(,nrow = nrow(ThirdSubstrateSet),ncol=1)
200 ITDAccessionNumbers<-matrix(data = Thirdsubbackfreq[1,],ncol = 1)
201
202 for (i in 1:nrow(ThirdSubstrateSet)){
203 ITDletters<-ThirdSubstrateSet[i,4:18]
204 ITDletters<-ITDletters[ITDletters !="XXXXX"]
205 ITDletters<-paste(ITDletters, sep="", collapse="")
206 YYYmotif <- unlist(strsplit(ITDletters, split = ""))
207 leftspaces<-c()
208 rightspaces<-c()
209 YYYposition <- match(x = "x", table = YYYmotif)
210 #position itself tells me how much is to the left of that X by what it's number is. x at position 4 tells me that there are
211 #just 3 letters to the left of x
212
213 YYYLettersToTheLeft <- YYYposition - 1
214 #how many letters to the right SHOULD just be length(motif)-position-1 if it's 5 long and x is at 3 then Y is at 4 and there is
215 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
216 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
217 #then sanity check, we're currently looking only at +/-4, but this spot allows for up to +/- 7 as well, just depends on what the
218 #variable the user puts in is
219 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
220 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
221 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
222 #add blank spaces if the motif has less than 4 letters to the left/right
223 motif<-c(leftspaces,YYYmotif,rightspaces)
224 #save that motif, which is the Y and +/- 4 amino acids, including truncation
225 motif<-motif[!motif %in% "x"]
226 motif<-paste(motif, sep="", collapse="")
227 ITDletters<-motif
228 ITDmotifs[i,1]<-ITDletters
229 # ITDAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
230 }
231
232 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
233 motif<-YYYmotif
234 #add blank spaces if the motif has less than 4 letters to the left/right
235 motif<-c(leftspaces,YYYmotif,rightspaces)
236 #save that motif, which is the Y and +/- 4 amino acids, including truncation
237 motif<-motif[!motif %in% "x"]
238 motif<-paste(motif, sep="", collapse="")
239 ITDletters<-motif
240 ITDmotifs[i,1]<-ITDletters
241 # ITDAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
242 }
243 }
244
245 #############################################################################################################################
246 #############################################################################################################################
247 #############################################################################################################################
248 #############################################################################################################################
249 #############################################################################################################################
250
251 #now look for either commonality or difference. Actually could you look for both...
252
253 if (Are_You_Looking_For_Commonality=="YES"){
254
255 columnalheader<-c(as.character(Thirdsubbackfreq[1:36,1]))
256 columnalheader<-matrix(columnalheader,nrow = 1)
257
258 SubstrateOverlap1<-intersect(D835Ymotifs,ITDmotifs)
259 SubstrateOverlap1<-as.matrix(SubstrateOverlap1)
260
261
262 columnalheader<-c(rep(NA,36))
263 FinalMatrix<-matrix(data =columnalheader,ncol = 1)
264
265 SubstrateOverlapFINAL<-intersect(FTLwtmotifs,SubstrateOverlap1)
266 AccessionOverlap1<-intersect(D835YAccessionNumbers,ITDAccessionNumbers)
267 AccessionOverlapFinal<-intersect(AccessionOverlap1,FTLwtAccessionNumbers)
268 AccessionOverlapFinal<-unlist(AccessionOverlapFinal)
269
270 for (x in 1:length(AccessionOverlapFinal)) {
271 for (y in 1:ncol(Firstsubbackfreq)) {
272 Acc<-AccessionOverlapFinal[x]
273 SBF<-Firstsubbackfreq[1,y]
274 if(Acc==SBF){
275 FinalMatrix<-cbind(FinalMatrix,Firstsubbackfreq[,y])
276 }
277 }
278 }
279 FinalMatrix<-FinalMatrix[,2:ncol(FinalMatrix)]
280
281 if(grepl(pattern = "Properties", x=FinalMatrix[22,1])==FALSE){
282 Outputmatrix<-cbind(Firstsubbackfreq[,1],FinalMatrix)
283 write.table(x=Outputmatrix,file = Shared_subbackfreq_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")
284 } else {
285 write.table(x=FinalMatrix,file = Shared_subbackfreq_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")
286 }
287
288 SubstrateMatrix<-SubstrateHeader
289 if(ncol(SubstrateMatrix)>18){
290 SubstrateMatrix<-SubstrateMatrix[,1:18]
291 }
292
293 for (z in 1:length(SubstrateOverlapFINAL)) {
294 motif<-SubstrateOverlapFINAL[z]
295 newmotif<-unlist(strsplit(motif,split = ""))
296
297 Addition<-""
298 outputmotif<-c(Addition,Addition,Addition,newmotif)
299 SubstrateMatrix<-rbind(SubstrateMatrix,outputmotif)
300 }
301 write.table(x=SubstrateMatrix,file = Shared_motifs_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")
302 }