Mercurial > repos > jfb > commonality_finder

comparison C and D finder/commonality-working jul 13 2020.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
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 #the above three sections bring in the input files and ensure they are properly aligned. The if statements align them if they are misaligned
32
33 First_unshared_motifs_table<-"R1 substrates.csv"
34 First_unshared_subbackfreq<-"R1 SBF.csv"
35
36 Second_unshared_motifs_table<-"R2 subs.csv"
37 Second_unshared_subbackfreq<-"R2 SBf.csv"
38
39 Third_unshared_motifs_table<-"R3 subs.csv"
40 Third_unshared_subbackfreq<-"R3 SBF.csv"
41
42 #the above 4 sections create the names of the output files that this tool can create
43
44
45
46
47 FirstxY<-rep("xY",times=nrow(FirstSubstrateSet))
48 FirstSubstrateSet[,11]<-FirstxY
49
50 SecondxY<-rep("xY",times=nrow(SecondSubstrateSet))
51 SecondSubstrateSet[,11]<-SecondxY
52
53 ThirdxY<-rep("xY",times=nrow(ThirdSubstrateSet))
54 ThirdSubstrateSet[,11]<-ThirdxY
55
56 #for each input file, mark the phospho-Tyrosine (which is always housed in column 11)
57 #with an xY to denote that is it a phospho and not regular tyrosine
58
59
60 #currently the substrates are a dataframe with many values, I want to collapse each substrate into a single variable,
61 #but if I simply use the paste() function then I lose any information about whether the substrate was truncated on the C or N terminal.
62 #so the for loops below then are constructed so as to retain that information on how any substrate is truncated.
63
64 FTLwtmotifs=matrix(,nrow = nrow(FirstSubstrateSet),ncol=1)
65 FTLwtAccessionNumbers=matrix(data = Firstsubbackfreq[1,],ncol=1)
66 #create the vectors which will house the first set of substrates and the first set of accession numbers
67
68
69 for (i in 1:nrow(FirstSubstrateSet)){
70 FTLwtletters<-FirstSubstrateSet[i,4:18]
71 FTLwtletters<-FTLwtletters[FTLwtletters !="XXXXX"]
72 FTLwtletters<-paste(FTLwtletters, sep="", collapse="")
73 leftspaces<-c()
74 rightspaces<-c()
75
76 YYYmotif <- unlist(strsplit(FTLwtletters, split = ""))
77 YYYposition <- match(x = "x", table = YYYmotif)
78 #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
79 #just 3 letters to the left of x
80
81 YYYLettersToTheLeft <- YYYposition - 1
82 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
83 #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
84 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
85
86
87
88 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
89 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
90 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
91 #add blank spaces if the motif has less than 4 letters to the left/right
92 motif<-c(leftspaces,YYYmotif,rightspaces)
93 #save that motif, which is the Y and +/- 4 amino acids, including truncation
94 motif<-motif[!motif %in% "x"]
95 motif<-paste(motif, sep="", collapse="")
96 FTLwtletters<-motif
97 FTLwtmotifs[i,1]<-FTLwtletters
98 }
99
100 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
101 motif<-YYYmotif
102 #add blank spaces if the motif has less than 4 letters to the left/right
103 motif<-c(leftspaces,YYYmotif,rightspaces)
104 #save that motif, which is the Y and +/- 4 amino acids, including truncation
105 motif<-motif[!motif %in% "x"]
106 motif<-paste(motif, sep="", collapse="")
107 FTLwtletters<-motif
108 FTLwtmotifs[i,1]<-FTLwtletters
109 }
110 }
111
112 D835Ymotifs=matrix(,nrow = nrow(SecondSubstrateSet),ncol=1)
113 D835YAccessionNumbers<-matrix(data = Secondsubbackfreq[1,],ncol = 1)
114 #vectors to house the second set of substrates and accession numbers
115
116 for (i in 1:nrow(SecondSubstrateSet)){
117 D835letters<-SecondSubstrateSet[i,4:18]
118 D835letters<-D835letters[D835letters !="XXXXX"]
119 D835letters<-paste(D835letters, sep="", collapse="")
120 leftspaces<-c()
121 rightspaces<-c()
122
123 YYYmotif <- unlist(strsplit(D835letters, split = ""))
124 YYYposition <- match(x = "x", table = YYYmotif)
125 #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
126 #just 3 letters to the left of x
127
128 YYYLettersToTheLeft <- YYYposition - 1
129 #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
130 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
131 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
132 #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
133 #variable the user puts in is
134 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
135 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
136 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
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 D835letters<-motif
143 D835Ymotifs[i,1]<-D835letters
144 }
145
146 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
147 motif<-YYYmotif
148 #add blank spaces if the motif has less than 4 letters to the left/right
149 motif<-c(leftspaces,YYYmotif,rightspaces)
150 #save that motif, which is the Y and +/- 4 amino acids, including truncation
151 motif<-motif[!motif %in% "x"]
152 motif<-paste(motif, sep="", collapse="")
153 D835letters<-motif
154 D835Ymotifs[i,1]<-D835letters
155 }
156 }
157
158
159 ITDmotifs=matrix(,nrow = nrow(ThirdSubstrateSet),ncol=1)
160 ITDAccessionNumbers<-matrix(data = Thirdsubbackfreq[1,],ncol = 1)
161 #third set of substrates and accession numbers
162
163
164 for (i in 1:nrow(ThirdSubstrateSet)){
165 ITDletters<-ThirdSubstrateSet[i,4:18]
166 ITDletters<-ITDletters[ITDletters !="XXXXX"]
167 ITDletters<-paste(ITDletters, sep="", collapse="")
168 YYYmotif <- unlist(strsplit(ITDletters, split = ""))
169 leftspaces<-c()
170 rightspaces<-c()
171 YYYposition <- match(x = "x", table = YYYmotif)
172 #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
173 #just 3 letters to the left of x
174
175 YYYLettersToTheLeft <- YYYposition - 1
176 #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
177 #just 1 spot to the right of Y so LettersToTheRight<-1 because 5-3-1=1
178 YYYLettersToTheRight <- length(YYYmotif) - YYYposition - 1
179 #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
180 #variable the user puts in is
181 if (YYYLettersToTheLeft < 7 | YYYLettersToTheRight < 7) {
182 leftspaces<-rep(" ",times=(7-YYYLettersToTheLeft))
183 rightspaces<-rep(" ",times=7-(YYYLettersToTheRight))
184 #add blank spaces if the motif has less than 4 letters to the left/right
185 motif<-c(leftspaces,YYYmotif,rightspaces)
186 #save that motif, which is the Y and +/- 4 amino acids, including truncation
187 motif<-motif[!motif %in% "x"]
188 motif<-paste(motif, sep="", collapse="")
189 ITDletters<-motif
190 ITDmotifs[i,1]<-ITDletters
191 # ITDAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
192 }
193
194 if(YYYLettersToTheLeft>6 && YYYLettersToTheRight>6){
195 motif<-YYYmotif
196 #add blank spaces if the motif has less than 4 letters to the left/right
197 motif<-c(leftspaces,YYYmotif,rightspaces)
198 #save that motif, which is the Y and +/- 4 amino acids, including truncation
199 motif<-motif[!motif %in% "x"]
200 motif<-paste(motif, sep="", collapse="")
201 ITDletters<-motif
202 ITDmotifs[i,1]<-ITDletters
203 # ITDAccessionNumbers[i,1]<-FirstSubstrateSet[i,3]
204 }
205 }
206
207
208 SubstrateOverlap1<-intersect(D835Ymotifs,ITDmotifs)
209 SubstrateOverlap1<-as.matrix(SubstrateOverlap1)
210 SubstrateOverlapFINAL<-intersect(FTLwtmotifs,SubstrateOverlap1)
211 #this tool is essentially looking for the intersection of three sets, and to do so it performs 1 intersection at a time
212 #this find the intersection of the substrate sets
213
214 AccessionOverlap1<-intersect(D835YAccessionNumbers,ITDAccessionNumbers)
215 AccessionOverlapFinal<-intersect(AccessionOverlap1,FTLwtAccessionNumbers)
216 AccessionOverlapFinal<-unlist(AccessionOverlapFinal)
217 #this tool is essentially looking for the intersection of three sets, and to do so it performs 1 intersection at a time
218 #this find the intersection of the accession number sets
219
220
221 for (x in 1:length(AccessionOverlapFinal)) {
222 for (y in 1:ncol(Firstsubbackfreq)) {
223 Acc<-AccessionOverlapFinal[x]
224 SBF<-Firstsubbackfreq[1,y]
225 if(Acc==SBF){
226 FinalMatrix<-cbind(FinalMatrix,Firstsubbackfreq[,y])
227 }
228 }
229 }
230 #for every accession number, go back to the original substrate background frequency file and find the protein statistics associated with that number
231 #columnbind all those protein statistics together
232 FinalMatrix<-FinalMatrix[,2:ncol(FinalMatrix)]
233
234
235 #write all files, in the proper format so that the files coming out look like the files coming in
236
237 if(grepl(pattern = "Properties", x=FinalMatrix[22,1])==FALSE){
238 Outputmatrix<-cbind(Firstsubbackfreq[,1],FinalMatrix)
239 write.table(x=Outputmatrix,file = Shared_subbackfreq_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")
240 } else {
241 write.table(x=FinalMatrix,file = Shared_subbackfreq_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")
242 }
243
244 SubstrateMatrix<-SubstrateHeader
245 if(ncol(SubstrateMatrix)>18){
246 SubstrateMatrix<-SubstrateMatrix[,1:18]
247 }
248
249 for (z in 1:length(SubstrateOverlapFINAL)) {
250 motif<-SubstrateOverlapFINAL[z]
251 newmotif<-unlist(strsplit(motif,split = ""))
252
253 Addition<-""
254 outputmotif<-c(Addition,Addition,Addition,newmotif)
255 SubstrateMatrix<-rbind(SubstrateMatrix,outputmotif)
256 }
257 write.table(x=SubstrateMatrix,file = Shared_motifs_table,quote = FALSE,sep = ",",row.names = FALSE,col.names = FALSE,na="")