|
4
|
1 NAMEOFOUTPUTFILE<-"output1.csv"
|
|
|
2
|
|
|
3 SuperAwesometrial <- read.delim2("input1.tabular", header=FALSE)
|
|
|
4 #once you've used the other script to turn the FASFA into a CSV, copypaste the filepath and name
|
|
|
5 #of the csv into this line between the quote marks.
|
|
|
6
|
|
|
7 SBF<-read.csv("input3.csv", stringsAsFactors = FALSE, header = FALSE)
|
|
|
8 SBF<-t(SBF)
|
|
|
9
|
|
|
10 PositiveMotifs <- read.csv("input2.csv", stringsAsFactors=FALSE)
|
|
|
11 #because of R reasons, it is required that the motifs in this file have blank cells instead of spaces where there is no letter in
|
|
|
12 #the motif
|
|
|
13
|
|
|
14 YsToim<-rep("xY",times=nrow(PositiveMotifs))
|
|
|
15 PositiveMotifs[,11]<-YsToim
|
|
|
16
|
|
|
17
|
|
|
18
|
|
|
19 ################################################################################################################################
|
|
|
20 #I have to paste them, then split and unlist them, then find the x and paste again
|
|
|
21 Positive9Letters<-PositiveMotifs[,4:18]
|
|
|
22 #head(Positive9Letters)
|
|
|
23 PositiveTrueMotifs<-c()
|
|
|
24
|
|
|
25 AccessionNumbers<-as.character(SBF[2:nrow(SBF),1])
|
|
|
26 AccessionNumbers<-AccessionNumbers[!is.na(AccessionNumbers)]
|
|
|
27 ALLPOSSIBLE<-SuperAwesometrial[,1]
|
|
|
28 ALLPOSSIBLE<-as.character(ALLPOSSIBLE)
|
|
|
29 ################################################################################################################################
|
|
|
30
|
|
|
31 for (q in 1:nrow(Positive9Letters)) {
|
|
|
32 LeftJust<-0
|
|
|
33 RightJust<-0
|
|
|
34
|
|
|
35 motifmotif<-Positive9Letters[q,]
|
|
|
36 motifmotif<-paste(motifmotif, collapse = "",sep = "")
|
|
|
37
|
|
|
38 motifmotif<-unlist(strsplit(motifmotif, split = ""))
|
|
|
39
|
|
|
40 position <- match(x = "x", table = motifmotif)
|
|
|
41 LeftJust<-position-1
|
|
|
42 RightJust<-length(motifmotif)-position-1
|
|
|
43
|
|
|
44 LeftSpaces<-rep(x=" ", times=(7-LeftJust))
|
|
|
45 RightSpaces<-rep(x=" ", times=(7-RightJust))
|
|
|
46
|
|
|
47 motifmotif<-motifmotif[!motifmotif %in% c("x")]
|
|
|
48
|
|
|
49 motifmotif<-c(LeftSpaces,motifmotif,RightSpaces)
|
|
|
50 motifmotif<-paste(motifmotif, collapse = "",sep = "")
|
|
|
51 PositiveTrueMotifs<-c(PositiveTrueMotifs,motifmotif)
|
|
|
52 }
|
|
|
53
|
|
|
54
|
|
|
55
|
|
|
56 ################################################################################################################################
|
|
|
57 allmotifs<-matrix(data=rep("Motifs", times= 1000000),ncol = 1)
|
|
|
58 thenames<-matrix(data=rep("AccessionNumbers", times= 1000000),ncol = 1)
|
|
|
59 ################################################################################################################################
|
|
|
60
|
|
|
61 ################################################################################################################################
|
|
|
62
|
|
|
63 #I need to preallocate these vectors. I will find out how many y's there are total and then make the vector that many long
|
|
|
64 #Or what I need is two separate loops. First loop finds all the accession number positions that Grep to the FASTA (which is called ALLPOSSIBLE)
|
|
|
65 #then take only those AAs from the fasta and count their y's, preallocate the vector for part 2 to that many y's
|
|
|
66 #those accessions and such as saved in a vector... this seems like it would be no faster actually
|
|
|
67
|
|
|
68 #then_that_are <- which(AccessionNumbers %in% ALLPOSSIBLE)
|
|
|
69
|
|
|
70 MotifNumber<-2
|
|
|
71
|
|
|
72 #TrueMotifNums<-which(ALLPOSSIBLE %in% AccessionNumbers)
|
|
|
73 #fihlodeANs<-c()
|
|
|
74
|
|
|
75 locations<-unique(grep(paste(AccessionNumbers,collapse="|"), ALLPOSSIBLE))
|
|
|
76
|
|
|
77 if (sum(locations)>0){
|
|
|
78 whereisit<-locations
|
|
|
79 for (u in 1:length(whereisit)) {
|
|
|
80 i<-whereisit[u]
|
|
|
81 name<-c()
|
|
|
82 data<-c()
|
|
|
83 name<-as.character(SuperAwesometrial[i,1])
|
|
|
84 #the name of each protein is the first column
|
|
|
85 name<-sub(x=name, pattern=",", replacement="")
|
|
|
86 #the names may contain commas, remove them
|
|
|
87 data<-as.character(SuperAwesometrial[i,3])
|
|
|
88 #the amino acids are stored in the third column
|
|
|
89 data<-strsplit(data,"")
|
|
|
90 #split them into their component letters
|
|
|
91 data<-unlist(data)
|
|
|
92 #turn them into a vector
|
|
|
93 motif<-c()
|
|
|
94
|
|
|
95 #this part below is where I can speed things up
|
|
|
96 The_Ys<-data=="Y"
|
|
|
97 #find any Y in the protein
|
|
|
98 if (sum(The_Ys>0)){ #if there is at least one Y
|
|
|
99 Where_are_they<-which(The_Ys %in% TRUE)
|
|
|
100 for (z in 1:length(Where_are_they)) { #then for every Y, make a motif
|
|
|
101
|
|
|
102 j<-Where_are_they[z]
|
|
|
103 #for (j in 1:length(data)){
|
|
|
104 #if ("Y" %in% data[j]){
|
|
|
105 #if there is a Y aka Tyrosine in the data
|
|
|
106 #allmotifs=rbind(allmotifs,data[(i-4):(i+4)])
|
|
|
107 a <- j-7
|
|
|
108 a<-ifelse(a<1, a <- 1, a <- a)
|
|
|
109 # if (a<1){
|
|
|
110 # a <- 1
|
|
|
111 # }
|
|
|
112 b<-j+7
|
|
|
113 b<-ifelse(b>length(data), b <- length(data), b <-
|
|
|
114 b)
|
|
|
115 # if (b>length(data)){
|
|
|
116 # b<-length(data)
|
|
|
117 # }
|
|
|
118 #take the motif that is +/- 4 from that Y, sanity checks so that values are never off the grid from the protein
|
|
|
119
|
|
|
120 LeftSide<-7-(j-a)
|
|
|
121 RightSide<-7-(b-j)
|
|
|
122 #how is the motif justified? Does it have exactly 4 letters to the left/right, or does it not?
|
|
|
123
|
|
|
124 leftspaces<-rep(" ",times=LeftSide)
|
|
|
125 rightspaces<-rep(" ",times=RightSide)
|
|
|
126 #add blank spaces if the motif has less than 4 letters to the left/right
|
|
|
127
|
|
|
128
|
|
|
129 motif<-(data[(a):(b)])
|
|
|
130 motif<-c(leftspaces,motif,rightspaces)
|
|
|
131 #save that motif, which is the Y and +/- 4 amino acids, including truncation
|
|
|
132
|
|
|
133 # lens<-c(lens,length(motif))
|
|
|
134 # leni<-c(leni,i)
|
|
|
135 # lenj<-c(lenj,j)
|
|
|
136
|
|
|
137 motif<-paste(motif, sep="", collapse="")
|
|
|
138 #the 4 amino acids, put them back together into a single string
|
|
|
139 motif<-matrix(data=c(motif),nrow = 1)
|
|
|
140 namesss<-matrix(data=c(name),nrow = 1)
|
|
|
141 #keep this motif and separately keep the name of the protein it came from
|
|
|
142
|
|
|
143 # allmotifs<-rbind(allmotifs,motif)
|
|
|
144 # thenames<-rbind(thenames,namesss)
|
|
|
145 allmotifs[MotifNumber,1]<-motif
|
|
|
146 thenames[MotifNumber,1]<-namesss
|
|
|
147 MotifNumber<-MotifNumber+1
|
|
|
148
|
|
|
149 #add names and motifs to a growing list
|
|
|
150
|
|
|
151 # write.table(motif, file="TRIALTIALRIAALSKFDJSD.csv", quote=FALSE, sep=",",
|
|
|
152 # row.names=FALSE,col.names = FALSE, na="", append=TRUE)
|
|
|
153 #and then write it into a csv, the sep is needed so that the two pieces of the data frame are separated
|
|
|
154 #append has 1to equal true because this thing will loop around many times adding more and more data points
|
|
|
155 #you must create a new filename/filepath with each new data you run
|
|
|
156 }
|
|
|
157
|
|
|
158 }
|
|
|
159 }
|
|
|
160 }
|
|
|
161
|
|
|
162
|
|
|
163
|
|
|
164
|
|
|
165 ################################################################################################################################
|
|
|
166 ################################################################################################################################
|
|
|
167 ################################################################################################################################
|
|
|
168
|
|
|
169
|
|
|
170 # for (i in 1:nrow(SuperAwesometrial)){
|
|
|
171 #
|
|
|
172 # }
|
|
|
173
|
|
|
174 names(allmotifs)<-thenames
|
|
|
175
|
|
|
176 truemotifs<-allmotifs[!duplicated(allmotifs)]
|
|
|
177 #truenames<-thenames[!duplicated(thenames)]
|
|
|
178 #remove duplicates from the motifs and names
|
|
|
179
|
|
|
180 #make the motifs and names into matrices
|
|
|
181
|
|
|
182
|
|
|
183 truemotifs<-truemotifs[!truemotifs %in% PositiveTrueMotifs]
|
|
|
184
|
|
|
185 outputfile<-cbind(names(truemotifs),truemotifs)
|
|
|
186
|
|
|
187 outputfile <- gsub(",","",outputfile)
|
|
|
188
|
|
|
189 write.table(outputfile, file=NAMEOFOUTPUTFILE, quote=FALSE, sep=",",
|
|
|
190 row.names=FALSE,col.names = FALSE, na="", append=TRUE)
|
