Upload to Gitlab UTU

0.ParseBamReadcount.R

+# To parse files generated by Bam-readcount
+
+# Details in field of bam read count
+
+#base:count:avg_mapping_quality:avg_basequality:avg_se_mapping_quality:num_plus_strand:num_minus_strand:avg_pos_as_fraction:avg_num_mismatches_as_fraction:avg_sum_mismatch_qualities:num_q2_containing_reads:avg_distance_to_q2_start_in_q2_reads:avg_clipped_length:avg_distance_to_effective_3p_end
+rm(list=ls())
+bam_rc.parse <- function(x){
+  columns = "base:count:avg_mapping_quality:avg_basequality:avg_se_mapping_quality:num_plus_strand:num_minus_strand:avg_pos_as_fraction:avg_num_mismatches_as_fraction:avg_sum_mismatch_qualities:num_q2_containing_reads:avg_distance_to_q2_start_in_q2_reads:avg_clipped_length:avg_distance_to_effective_3p_end"
+  column.names = strsplit(columns, ":")[[1]];rm(columns)
+  # get line
+  Row = strsplit(x, split = "\t")[[1]]
+  # get base counts
+  temp <- lapply(Row[-c(1:4)], function(basecnt){
+    basecnt = strsplit(basecnt, split = ":")[[1]]
+  }) %>% do.call(rbind, .)
+  
+  colnames(temp) = column.names
+  
+  dat = data.frame(chr =Row[1],pos = Row[2],ref = Row[3], dp = Row[4],
+                   temp)
+  
+}
+strandBias=function(arg1,arg2){
+  if(length(arg1)!=length(arg2)){
+    error=paste("Parameters with unequal lengths passed as argument!\narg1 =",length(arg1),"; arg2 =",length(arg2))
+    message(error)
+    return(NA)
+  }
+  SB=c()
+  for (i in seq(1,length(arg1))){
+    if(is.na(arg1[i]) | is.na(arg2[i])){
+      calc=NA
+    }
+    else if(arg1[i]<arg2[i]){
+      calc=(arg2[i]/arg1[i])*-1
+    }
+    else{
+      calc=(arg1[i]/arg2[i])
+    }
+    SB=c(SB,calc)
+  }
+  return(SB)
+}
+
+library(dplyr)
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/Bam-readcount")
+
+files=list.files(".",pattern = "bamcount")
+for (file in files){
+  #file="EGFR.Lib.S1.bam.tsv"
+  #file="EGFREGFP5.S1.bam.tsv"
+  #file="EGFREGFP1.S1.bam.tsv"
+  temp=scan(file, what = "character", sep = "\n")
+  dat = lapply(temp, bam_rc.parse) %>% bind_rows()
+  dat$chr=as.character(dat$chr)
+  dat$pos=as.numeric(dat$pos)
+  dat$dp=as.numeric(dat$dp)
+  dat$count=as.numeric(dat$count)
+  dat[,7:18]=sapply(dat[,7:18],as.numeric)
+  dat=dat[dat$base!="=",]
+  dat$strandbias=strandBias(dat$num_plus_strand,dat$num_minus_strand)
+  dat$AlleleFreq=dat$count/dat$dp
+  saveRDS(dat,file=paste("Parsed/parsed.",gsub(".tsv","",file),".RDS",sep=""))
+  write.table(dat,file=paste("Parsed/parsed.",file,sep=""),sep="\t",row.names = F,col.names = T,quote = F,na = "NA")
+  
+}
+

1.ParseVCF.R

+# To parse VCF files generated from samtools mpileup & bcftools
+rm(list=ls())
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf")
+# file="20180112 EGFR_lib.EGFR.locus.HiDP.vcf"
+file="20180112 EGFR.No.Lig.EGFR.locus.HiDP.vcf"
+
+temptext=readLines(paste(file,sep=""))
+temptext=temptext[-grep("^##",x = temptext)]
+require(progress)
+pb <- progress_bar$new( format = "  Processing [:bar] :percent in :elapsed", total = length(temptext), clear = FALSE, width= 60)
+temptext=gsub(";","\t",temptext)
+temptext=temptext[-seq(1:49)]
+parsing=matrix(NA,length(temptext),19)
+colnames(parsing)=c("chr","pos","dot1","Ref","Alt","zero","dot2","DP","I16","QS","VDB","SGB","RPB","MQB","MQSB","BQB","MQ0F","Format","DATA")
+for( i in seq(1,length(temptext))){
+#for( i in seq(69801,69810)){
+  line=temptext[i]
+  words=unlist(strsplit(line,"\t"))
+  parsing[i,1:10]=words[1:10]
+  for (j in seq(11,19)){
+    if(is.na(words[j])|words[j]==""){
+      break
+    }
+    #print(regexpr("VDB",words[j]))
+    if(regexpr("VDB",words[j])==1){
+      parsing[i,"VDB"]=words[j]
+    }
+    else if(regexpr("SGB",words[j])==1){
+      parsing[i,"SGB"]=words[j]
+    }
+    else if(regexpr("RPB",words[j])==1){
+      parsing[i,"RPB"]=words[j]
+    }
+    else if(regexpr("MQB",words[j])==1){
+      parsing[i,"MQB"]=words[j]
+    }
+    else if(regexpr("MQSB",words[j])==1){
+      parsing[i,"MQSB"]=words[j]
+    }
+    else if(regexpr("BQB",words[j])==1){
+      parsing[i,"BQB"]=words[j]
+    }
+    else if(regexpr("MQ0F",words[j])==1){
+      parsing[i,"MQ0F"]=words[j]
+    }
+    else if(regexpr("PL:AD",words[j])==1){
+      parsing[i,"Format"]=words[j]
+      parsing[i,"DATA"]=words[j+1]
+    }
+  }
+  if(i%%1000==0){
+    pb$tick(1000)
+  }
+}
+temp=data.frame(parsing,stringsAsFactors = F)
+
+pb <- progress_bar$new( format = "  Processing [:bar] :percent in :elapsed", total = length(temp$DATA), clear = FALSE, width= 60)
+PL=c();AD=c()
+for(i in seq(1,length(temp$DATA))){
+#for(i in seq(1,10)){
+  #print(temp$DATA[i])
+  var=unlist(strsplit(temp$DATA[i],":"))
+  PL=c(PL,var[1])
+  AD=c(AD,var[2])
+  if(i%%1000==0){
+    pb$tick(1000)
+  }
+}
+temp=cbind(temp,PL,AD)
+temp=temp[,c(-18,-19)]
+colnames(temp)[c(19)]="Allelic.Depth"
+parsing=temp;rm(temp)
+
+assign(file,parsing);rm(parsing)
+file.s=paste("Parsed/P-",file,sep="")
+saveRDS(object =get(file),file = paste(file.s,".RDS",sep=""))
+write.table(get(file),file=paste(file.s,".tsv",sep=""),sep="\t",row.names = F,col.names = T,quote = F,na = "NA")
+
+

2.FilterVCF.R

+bdir="~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/Parsed/";setwd(bdir)
+
+# rm(list=ls());file="P-20180112 EGFR_lib.EGFR.locus.HiDP.vcf.tsv"
+rm(list=ls());file="P-20180112 EGFR.No.Lig.EGFR.locus.HiDP.vcf.tsv"
+
+#setwd(paste(bdir,dir,sep=""))
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+tab=read.table(file,sep="\t",header = T,as.is = T,stringsAsFactors = F)
+tab=tab[c(-3,-6,-7)] # Removing zero and dot (.) columns
+#Change Depth to numeric
+tab$DP=as.numeric(gsub("DP=","",tab$DP))
+
+# Subsetting to EGFR locus #chr7:55,084,725-55,277,031
+subtab=subset(tab,tab$chr=="chr7" & tab$pos>55084725 & tab$pos < 55277031)
+
+#Subsetting to get Depth > 1000
+# subtab=subset(subtab,subtab$DP>1000)
+
+# Make New Only Mutation Matrix
+
+require(progress);pb <- progress_bar$new( format = "  Processing [:bar] :current SNVs processed in - :elapsed", total = length(unlist(strsplit(subtab$Alt,","))), clear = FALSE, width= 65)
+chr=c();pos=c();ref=c();alt=c();DP=c();AD=c();count=0
+for(i in seq(1,length(subtab$DP))){
+#for(i in seq(1,100)){
+  datarow=subtab[i,]
+  Alts=unlist(strsplit(datarow[,"Alt"],",",fixed = T),use.names = F)
+  ADs=unlist(strsplit(datarow[,"Allelic.Depth"],",",fixed = T),use.names = T)
+  for(j in seq(1,length(Alts))){
+    count=count+1
+    #chr=unlist(c(chr,datarow[1]))
+    chr=append(chr,unlist(datarow[1]))
+    pos=append(pos,unlist(datarow[2]))
+    ref=append(ref,unlist(datarow[3]))
+    alt=append(alt,Alts[j])
+    DP=append(DP,unlist(datarow[5]))
+    AD=append(AD,ADs[j+1])
+    if(count%%100==0){
+      pb$tick(100)
+    }
+  }
+}
+AD=as.numeric(AD)
+SNV.matrix=data.frame(chr,pos,ref,alt,DP,AD,stringsAsFactors = F)
+rm(chr,pos,ref,alt,DP,AD,Alts,ADs,datarow,i,j)
+submat=subset(SNV.matrix,SNV.matrix$alt!="<*>")
+
+fnm=gsub("P-","../Variants/V-",gsub(".vcf.tsv","",file))
+saveRDS(object = submat,file = paste(fnm,".RDS",sep="")) # use var=readRDS("FileName")
+write.table(submat,file = paste(fnm,".tsv",sep=""),sep="\t",row.names = F,col.names = T,quote = F,na = "NA")
+
+

3.FoldChange & Plots.R

+
+#Subsetting to get Depth > 1000
+#subtab=subset(subtab,subtab$DP>1000)
+
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/Variants/")
+
+rm(list=ls())
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+
+
+
+file1="V-20180112 EGFR_lib.EGFR.locus.HiDP.RDS";tab=readRDS(file1)
+VF=(tab$AD*100)/tab$DP
+Rel.Ab=(tab$AD*100)/sum(tab$AD)
+MutID=paste(tab$chr,":",tab$pos,tab$ref,">",tab$alt,sep="")
+tab=data.frame(tab,VF,Rel.Ab,MutID,stringsAsFactors = F)
+rm(VF,Rel.Ab,MutID)
+Library=tab;rm(tab,file1)
+colnames(Library)=paste("Lib.",colnames(Library),sep="")
+
+file1="V-20180112 EGFR.No.Lig.EGFR.locus.HiDP.RDS";tab=readRDS(file1)
+VF=(tab$AD*100)/tab$DP
+Rel.Ab=(tab$AD*100)/sum(tab$AD)
+MutID=paste(tab$chr,":",tab$pos,tab$ref,">",tab$alt,sep="")
+tab=data.frame(tab,VF,Rel.Ab,MutID,stringsAsFactors = F)
+rm(VF,Rel.Ab,MutID)
+NoLigand=tab;rm(tab,file1)
+colnames(NoLigand)=paste("NoLig.",colnames(NoLigand),sep="")
+
+allMuts=union(NoLigand$NoLig.MutID,Library$Lib.MutID)
+idx=match(allMuts,Library$Lib.MutID)
+Mutation.Table=Library[idx,]
+idx=match(allMuts,NoLigand$NoLig.MutID)
+Mutation.Table=data.frame(allMuts,Mutation.Table,NoLigand[idx,],stringsAsFactors = F)
+rm(Library,NoLigand)
+
+FoldChange=function(arg1,arg2){
+  if(length(arg1)!=length(arg2)){
+    error=paste("Parameters with unequal lengths passed as argument!\narg1 =",length(arg1),"; arg2 =",length(arg2))
+    message(error)
+    return(NA)
+  }
+  FC=c()
+  for (i in seq(1,length(arg1))){
+    if(is.na(arg1[i]) | is.na(arg2[i])){
+      calc=NA
+    }
+    else if(arg1[i]<arg2[i]){
+      calc=(arg2[i]/arg1[i])
+    }
+    else{
+      calc=(arg1[i]/arg2[i])*-1
+    }
+    FC=c(FC,calc)
+  }
+  return(FC)
+}
+
+Mutation.Table$FC=FoldChange(Mutation.Table$Lib.VF,Mutation.Table$NoLig.VF)
+
+AnnoDF=Mutation.Table[,c(2,3,3,4,5,1)]
+# write.table(AnnoDF,file = "../20180131.InputAnnovar.txt",col.names = F,quote = F,row.names = F)
+
+#- run following line in annovar
+# annotate_variation.pl -out EGFR.muts --build hg19 20180131.InputAnnovar.txt humandb
+#- more annotations
+# table_annovar.pl "20180131.InputAnnovar.txt" ~/NGS_Seq_Tools/annovar2/humandb -buildver hg19 -out EGFR.iSCREAM -remove -protocol refgene,gnomad_genome,gnomad_exome,avsnp150,clinvar_20170905 -operation g,f,f,f,f -nastring . -csvout -polish
+rm(AnnoDF)
+var=read.table("../EGFR.iSCREAM.hg19_multianno.csv",sep=",",stringsAsFactors = F,header = T)
+var$MutID=paste(var$Chr,":",var$Start,var$Ref,">",var$Alt,sep="")
+
+idx=match(Mutation.Table$allMuts,var$MutID)
+Mutation.Table=data.frame(Mutation.Table,var[idx,c(6:26)])
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/EleniusGroup/DvsP.EGFR.HiDP/AnnovarMutCodeFind.R")
+
+Mutation.Table$AAchange=annovarMutCodeFind(Mutation.Table$AAChange.refgene,isoform = "NM_005228")
+temp=gsub("^.","",x = Mutation.Table$AAchange);temp=gsub(".$","",x=temp)
+Mutation.Table$AAPos=as.numeric(temp);rm(temp)
+
+Mutation.Table=Mutation.Table[Mutation.Table$ExonicFunc.refgene!="synonymous SNV",]
+
+# Add BamReadCount
+rm(var,idx)
+Lib.Count=readRDS("../../Bam-readcount/Parsed/parsed.EGFR.Lib.bamcount.RDS")
+Lib.Count$MutID=paste(Lib.Count$chr,":",Lib.Count$pos,Lib.Count$ref,">",Lib.Count$base,sep="")
+colnames(Lib.Count)=paste("bamRC.Lib.",colnames(Lib.Count),sep="")
+idx=match(Mutation.Table$allMuts,Lib.Count$bamRC.Lib.MutID)
+test=cbind.data.frame(Mutation.Table,Lib.Count[idx,])
+
+NoLig.Count=readRDS("../../Bam-readcount/Parsed/parsed.EGFR.NoLig.bamcount.RDS")
+NoLig.Count$MutID=paste(NoLig.Count$chr,":",NoLig.Count$pos,NoLig.Count$ref,">",NoLig.Count$base,sep="")
+colnames(NoLig.Count)=paste("bamRC.NoLig.",colnames(NoLig.Count),sep="")
+idx=match(Mutation.Table$allMuts,NoLig.Count$bamRC.NoLig.MutID)
+test=cbind.data.frame(test,NoLig.Count[idx,])
+
+Mutation.Table=test;rm(test,Lib.Count,NoLig.Count,idx)
+
+# write.table(Mutation.Table,file = "../20180206.Result.tsv",sep="\t",col.names = T,quote = F,row.names = F)
+# saveRDS(Mutation.Table,file = "../20180206.Result.RDS")
+
+
+#------------------------- 
+library(ggplot2)
+source("~/Seafile/My Library/Git/GitLab.UTU/Misc Code/Master ggplot2 Theme/theme.DC.plot.R")
+# All data. Massive plot!!!
+FC.data.cut=droplevels.data.frame(na.exclude(Mutation.Table))
+# View(FC.data.cut[FC.data.cut$AAchange%in%c("L858R","T790M","A702V","A1118E","L184P"),])
+
+# str.extra(FC.data.cut)
+toRemove=c(10,11,12,13,14,19,21,23,26,28,29,30,31,32,33,34,36,38,39,40,seq(44,46),48,seq(50,52),seq(55,61),64,seq(65,67),69,seq(71,73),seq(76,82),85)
+colnames(FC.data.cut)[toRemove]
+
+FC.data.cut=FC.data.cut[,-toRemove]
+# Removed columns
+# [1] "Lib.MutID"          "NoLig.chr"         
+# [3] "NoLig.pos"          "NoLig.ref"         
+# [5] "NoLig.alt"          "NoLig.MutID"       
+# [7] "Func.refgene"       "GeneDetail.refgene"
+# [9] "Xref.refgene"       "gnomAD_genome_AFR" 
+# [11] "gnomAD_genome_AMR"  "gnomAD_genome_ASJ" 
+# [13] "gnomAD_genome_EAS"  "gnomAD_genome_FIN" 
+# [15] "gnomAD_genome_NFE"  "gnomAD_genome_OTH" 
+# [17] "CLINSIG"            "CLNACC"            
+# [19] "CLNDSDB"            "CLNDSDBID"         
+# [21] "MutID"       
+
+FC.data.cut$Direction=FC.data.cut$FC
+FC.data.cut$Direction[FC.data.cut$Direction<0]="Decrease"
+FC.data.cut$Direction[FC.data.cut$Direction!="Decrease"]="Increase"
+FC.data.cut$Direction=as.factor(FC.data.cut$Direction)
+
+ggplot(data = FC.data.cut,aes(x=bamRC.NoLig.strandbias,y=FC.data.cut$bamRC.Lib.strandbias))+geom_point(alpha=0.4,size=2)+theme_dc.plot.xaxis.regular+geom_vline(xintercept = 0)+geom_hline(yintercept = 0)+xlab("StrandBias NoLigand")+ylab("Strand Bias Plasmid Library")+ggtitle("Strand Bias distribution")+ggtitle(paste("Strand Bias distribution N=",length(FC.data.cut$allMuts)))
+
+# removing infinite strand bias AA changes
+FC.data.cut=FC.data.cut[!((FC.data.cut$bamRC.NoLig.strandbias==Inf | FC.data.cut$bamRC.NoLig.strandbias==-Inf)|(FC.data.cut$bamRC.Lib.strandbias==Inf | FC.data.cut$bamRC.Lib.strandbias==-Inf)),]
+                        
+ggplot(data = FC.data.cut,aes(x=bamRC.NoLig.strandbias,y=FC.data.cut$bamRC.Lib.strandbias))+geom_point(alpha=0.4,size=2)+theme_dc.plot.xaxis.regular+geom_vline(xintercept = 0)+geom_hline(yintercept = 0)+xlab("StrandBias NoLigand")+ylab("Strand Bias Plasmid Library")+ggtitle("Strand Bias distribution (Inf removed)")+ggtitle(paste("Strand Bias distribution N=",length(FC.data.cut$allMuts)))
+
+
+FC.data.cut=FC.data.cut[FC.data.cut$bamRC.NoLig.strandbias<=10&FC.data.cut$bamRC.NoLig.strandbias>=-10|FC.data.cut$bamRC.Lib.strandbias<=10&FC.data.cut$bamRC.Lib.strandbias>=-10,]
+
+ggplot(data = FC.data.cut,aes(x=bamRC.NoLig.strandbias,y=FC.data.cut$bamRC.Lib.strandbias))+geom_point(alpha=0.4,size=2)+theme_dc.plot.xaxis.regular+geom_vline(xintercept = 0)+geom_hline(yintercept = 0)+xlab("StrandBias NoLigand")+ylab("Strand Bias Plasmid Library")+ggtitle(paste("Strand Bias distribution N=",length(FC.data.cut$allMuts)))
+
+
+# write.table(FC.data.cut,file = "../20180206.Final.Table.tsv",sep="\t",col.names = T,quote = F,row.names = F)
+# saveRDS(FC.data.cut,file = "../20180206.Final.table.RDS")
+
+library(RColorBrewer)
+cols=colorRampPalette(c("#003366","firebrick1"))(20)
+#cols=colorRampPalette(c("#ffeaea","#ff3030"))(20)
+
+# #temp=temp[order(temp$VF.NoLigand,decreasing = F),]
+# p=ggplot(FC.data.cut,aes(y=FC,x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.9)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC>25,"AAchange"]),hjust=1.3,size=3)+scale_y_continuous(breaks=seq(-240,100,by = 20))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC<(-50),],aes(label=FC.data.cut[FC.data.cut$FC<(-50),"AAchange"]),hjust=1.3,size=3)
+# pdf(file = "../FC.plot.pdf",width = 10,height = 7);print(p);dev.off()
+# 
+# p=ggplot(FC.data.cut,aes(y=FC,x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.1)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC>25,"AAchange"]),hjust=1.3,size=3)+scale_y_continuous(breaks=seq(-240,100,by = 20))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC<(-50),],aes(label=FC.data.cut[FC.data.cut$FC<(-50),"AAchange"]),hjust=1.3,size=3)
+# pdf(file = "../FC.plot.alpha.pdf",width = 10,height = 7);print(p);dev.off()
+# ggplotly(p)
+
+# p=ggplot(FC.data.cut,aes(y=FC,x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.9)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC>25,"AAchange"]),hjust=1.3,size=3)+scale_y_continuous(breaks=seq(-240,100,by = 20))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC<(-50),],aes(label=FC.data.cut[FC.data.cut$FC<(-50),"AAchange"]),hjust=1.3,size=3)+geom_rug(sides="r",alpha=0.5,size=0.2,color="black");p
+# ggsave(plot = p,filename = "../FC.plot.w.rug.pdf",width = 10,height = 7);print(p);dev.off()
+
+
+# Only FC not my signed FC.
+
+FC.data.cut$FC.reg=(FC.data.cut$NoLig.VF/FC.data.cut$Lib.VF)
+p=ggplot(FC.data.cut,aes(y=FC.reg,x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.9)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC.reg>25,"AAchange"]),hjust=1.3,size=3)+geom_hline(yintercept = 0,color="black")+geom_hline(yintercept = 1,linetype="dashed",color="red",alpha=0.5);p
+pdf(file = "../FCreg.plot.pdf",width = 10,height = 5);print(p);dev.off()
+# 
+# p=ggplot(FC.data.cut,aes(y=FC.reg,x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.1)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC.reg>25,"AAchange"]),hjust=1.3,size=3)+geom_hline(yintercept = 0,color="black")+geom_hline(yintercept = 1,linetype="dashed",color="red",alpha=0.5);p
+# pdf(file = "../FCreg.alpha.plot.pdf",width = 10,height = 5);print(p);dev.off()
+
+
+# base=c(1,2,4,6,8)
+# y.breaks=sort(c(0.01,0.1,base*1,base*10,100))
+# 
+# p=ggplot(FC.data.cut,aes(y=log2(FC.reg),x=AAPos,size=NoLig.VF,color=NoLig.VF))+geom_point(alpha=0.9)+scale_colour_gradient(low="#003366",high="firebrick1")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("log2(Fold Change)")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC.reg>25,"AAchange"]),hjust=1.3,size=3)+scale_y_continuous(breaks = log2(y.breaks),labels=c(0.01,0.1,1,rep("",4),10,rep("",4),100))+geom_hline(yintercept = 0,color="black");p
+# pdf(file = "../FCreg.log.plot.pdf",width = 10,height = 7);print(p);dev.off()
+
+
+# ggplotly(p)
+
+# Browsable plot
+mini=FC.data.cut[,c("AAchange","Lib.VF","NoLig.VF","Lib.AD","NoLig.AD","FC.reg","AAPos")]
+colnames(mini)[c(3)]=c("VariantFrequency")
+
+library(plotly)
+
+X.A <- list(
+  title = "EGFR amino acid",
+  showticklabels = TRUE,
+  dtick=100,
+  ticklen = 5,
+  gridcolor = toRGB("white"),
+  gridwidth = 0.5,
+  tickwidth = 1,
+  range = c(-10, 1220),
+  tickangle = 0,
+  zerolinecolor = toRGB("black"),
+  zerolinewidth = 1.5)
+
+Y.A <- list(
+  title = "Fold Change",
+  showticklabels = TRUE,
+  dtick=25,
+  ticklen = 5,
+  gridcolor = toRGB("white"),
+  gridwidth = 0.5,
+  tickwidth = 1,
+  zerolinecolor = toRGB("black"),
+  zerolinewidth = 1.5)
+
+p=plot_ly(mini,alpha=0.7,x=~AAPos,y=~FC.reg,color=~VariantFrequency,size=~VariantFrequency,mode="text",text=~paste('EGFR ',AAchange,'</br> FC: ',round(FC.reg,digits = 3),'</br>- - - -</br>Variant frequency</br>Library: ',round(Lib.VF,digits = 3),'</br>Surving cells: ',round(VariantFrequency,digits=3),'</br>- - - -</br>Number of reads</br>Library: ',Lib.AD,'</br>Surving cells: ',NoLig.AD),colors = cols,sizes=c(50,300))%>%
+  add_markers()%>%
+  layout(title = "EGFR iSREAM",xaxis = X.A,yaxis = Y.A)
+print(p)
+# 
+# 
+# library(htmlwidgets)
+# library(htmltools)
+# 
+# p <- htmlwidgets::appendContent(p, htmltools::tags$input(id='inputText', value='L858R', ''), htmltools::tags$button(id='buttonSearch', 'Search'))
+# p <- htmlwidgets::appendContent(p, htmltools::tags$script(HTML(
+#   'document.getElementById("buttonSearch").addEventListener("click", function()
+#   {        
+#   var i = 0;
+#   var j = 0;
+#   var found = [];
+#   var myDiv = document.getElementsByClassName("js-plotly-plot")[0]
+#   var data = JSON.parse(document.querySelectorAll("script[type=\'application/json\']")[0].innerHTML);
+#   for (i = 0 ;i < data.x.data.length; i += 1) {
+#   for (j = 0; j < data.x.data[i].text.length; j += 1) {
+#   if (data.x.data[i].text[j].indexOf(document.getElementById("inputText").value) !== -1) {
+#   found.push({curveNumber: i, pointNumber: j});
+#   }
+#   }
+#   }
+#   Plotly.Fx.hover(myDiv, found);
+#   }  
+# );')))                                                    
+# 
+# htmlwidgets::saveWidget(p, paste('../FoldChangeRegulariSCREAM', ".html", sep=""))
+# p
+# 
+# # ,text=~paste('EGFR Mut: ',AAchange,'</br> Reads_Library: ',Lib.AD,'</br> VF_Lib: ',round(Lib.VF,digits = 3),'</br> Reads_NoLig: ',NoLig.AD,'</br> VF_NoLig: ',round(NoLig.VF,digits=3),'</br> FC: ',round(FC.reg,digits = 3))

3b.Animate Foldchange plot.R

+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/Variants/")
+
+rm(list=ls())
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+
+FC.data.cut=readRDS("../20180401.MutationTable.RDS")
+FC.data.cut$FC.reg=(FC.data.cut$NoLig.VF/FC.data.cut$Lib.VF)
+
+selected=FC.data.cut[,c("AAchange","AAPos","Lib.VF","NoLig.VF","FC.reg")]
+
+expandFrames=function(FC,VF.Begin,VF.End,frames=30){
+  # FC is foldchange column
+  # frames is the number of frames you want for a smooth animation
+  DF=matrix(data = 0,nrow =length(FC),ncol = frames)
+  VF=matrix(data = 0,nrow =length(VF.Begin),ncol = frames)
+  for(i in seq(1,length(FC))){
+    DF[i,]=seq(from = 0,to = FC[i],length.out = frames)
+    VF[i,]=seq(from = VF.Begin[i],to = VF.End[i],length.out = frames)
+  }
+  DF=as.data.frame(DF)
+  colnames(DF)=paste("FC.frame.",seq(1,frames,by = 1),sep="")
+  VF=as.data.frame(VF)
+  colnames(VF)=paste("VF.frame.",seq(1,frames,by = 1),sep="")
+  out.DF=cbind(DF,VF)
+  return(out.DF)
+}
+
+DF=expandFrames(FC = selected$FC,VF.Begin = selected$Lib.VF,VF.End = selected$NoLig.VF,frames = 60)
+300
+# setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/Variants/")
+# 
+# rm(list=ls())
+# source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+# # saveRDS(Mutation.Table,file = "../20180206.Result.RDS")
+# 
+# FC.data.cut=readRDS("../20180206.Final.table.RDS")
+# selected=FC.data.cut[,c("AAchange","AAPos","Lib.VF","NoLig.VF","FC")]
+# 
+# expandFrames=function(FC,VF.Begin,VF.End,frames=30){
+#   # FC is foldchange column
+#   # frames is the number of frames you want for a smooth animation
+#   DF=matrix(data = 0,nrow =length(FC),ncol = frames)
+#   VF=matrix(data = 0,nrow =length(VF.Begin),ncol = frames)
+#   for(i in seq(1,length(FC))){
+#     DF[i,]=seq(from = 0,to = FC[i],length.out = frames)
+#     VF[i,]=seq(from = VF.Begin[i],to = VF.End[i],length.out = frames)
+#   }
+#   DF=as.data.frame(DF)
+#   colnames(DF)=paste("FC.frame.",seq(1,frames,by = 1),sep="")
+#   VF=as.data.frame(VF)
+#   colnames(VF)=paste("VF.frame.",seq(1,frames,by = 1),sep="")
+#   out.DF=cbind(DF,VF)
+#   return(out.DF)
+# }
+# 
+# DF=expandFrames(FC = selected$FC,VF.Begin = selected$Lib.VF,VF.End = selected$NoLig.VF,frames = 60)
+# selected=cbind(selected,DF)
+# # FC = selected$FC;VF.Begin = selected$Lib.VF;VF.End = selected$NoLig.VF;frames = 20
+# 
+# FC.data=selected[,1:5];colnames(FC.data)
+# # [1] "AAchange" "AAPos"    "Lib.VF"   "NoLig.VF" "FC"    
+# FC.data$FC=rep(0,length(FC.data$FC))
+# FC.data$VF=rep(0,length(FC.data$Lib.VF))
+# 
+# frames=grep("FC.frame",colnames(selected))
+# VF.frames=grep("VF.frame",colnames(selected))
+# j=1
+# VF.range=seq(0,5,by = 0.01)
+# color.range=colorRampPalette(c("#003366","firebrick1","firebrick1","firebrick1","firebrick1","firebrick1","firebrick1"))(length(VF.range))
+# for(i in frames){
+#   print(paste("Frame",i))
+#   FC.data$FC=selected[,i]
+#   FC.data$VF=selected[,VF.frames[j]];j=j+1
+#   p=ggplot(FC.data,aes(y=FC,x=AAPos,size=VF,color=VF))+geom_point(alpha=0.9)+scale_colour_gradientn(colours = color.range,values=VF.range)+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_hline(yintercept = 0,color="black")+scale_y_continuous(breaks=seq(-240,100,by = 20),limits = c(-240,110))+geom_hline(yintercept = 0,color="black")
+#  ggsave(filename = paste("../Animated.FC/FC.plot.frame",stringr::str_pad(i,2,pad = "0"),".png",sep = ""),plot = p,dpi = 200,width = 10,height = 7)
+# };j=1
+# # Final frame
+# FC.data.cut$VF=FC.data.cut$NoLig.VF
+# p=ggplot(FC.data.cut,aes(y=FC,x=AAPos,size=VF,color=VF))+geom_point(alpha=0.9)+scale_colour_gradientn(colours = color.range,values=VF.range)+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text.y= element_text(size = rel(1),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"),axis.text.x = element_text(angle = 90, hjust = 1,size = rel(0.75)))+ggtitle("Fold Change Map of Mutations (From Library to Ligand Independent)")+xlab("Amino Acid Position")+ylab("Fold Change")+scale_x_continuous(breaks = c(1,seq(50,1210,by = 50),1210))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC>25,],aes(label=FC.data.cut[FC.data.cut$FC>25,"AAchange"]),hjust=1.3,size=3)+scale_y_continuous(breaks=seq(-240,100,by = 20),limits = c(-240,110))+geom_hline(yintercept = 0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FC<(-50),],aes(label=FC.data.cut[FC.data.cut$FC<(-50),"AAchange"]),hjust=1.3,size=3)
+# ggsave(filename = paste("../Animated.FC/FC.plot.frame",paste("../Animated.FC/FC.plot.frame",stringr::str_pad(i+1,2,pad = "0"),".png",sep = ""),".png",sep = ""),plot = p,dpi = 200,width = 10,height = 7)
+# 
+# 
+# # run this: convert -delay 0.1 -loop 1 *.png animation.gif
+# 

4.SankeyStats.R

+
+load("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/EleniusGroup/ERBB.Miscellaneous/Posssible mutants/EGFR/egfr_Variants.bin")
+print(paste("Unique EGFR cDNA=",length(unique(Variants$mutants)),"Unique EGFR proteins=",length(unique(Variants$mutantprotein))))
+
+EGFR.WT=Variants$mutantprotein[1]
+non.syn=Variants[Variants$mutantprotein!=EGFR.WT,]
+length(unique(non.syn$mutants))+1 # +1 for WT
+length(unique(non.syn$mutantprotein))+1 # +1 for WT
+
+print(paste("Non-synonymous EGFR muts =",length(non.syn$mutants)))
+
+#Subsetting to get Depth > 1000
+#subtab=subset(subtab,subtab$DP>1000)
+
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/Variants/")
+
+rm(list=ls())
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+
+
+
+file1="V-20180112 EGFR_lib.EGFR.locus.HiDP.RDS";tab=readRDS(file1)
+VF=(tab$AD*100)/tab$DP
+Rel.Ab=(tab$AD*100)/sum(tab$AD)
+MutID=paste(tab$chr,":",tab$pos,tab$ref,">",tab$alt,sep="")
+tab=data.frame(tab,VF,Rel.Ab,MutID,stringsAsFactors = F)
+rm(VF,Rel.Ab,MutID)
+Library=tab;rm(tab,file1)
+colnames(Library)=paste("Lib.",colnames(Library),sep="")
+
+file1="V-20180112 EGFR.No.Lig.EGFR.locus.HiDP.RDS";tab=readRDS(file1)
+VF=(tab$AD*100)/tab$DP
+Rel.Ab=(tab$AD*100)/sum(tab$AD)
+MutID=paste(tab$chr,":",tab$pos,tab$ref,">",tab$alt,sep="")
+tab=data.frame(tab,VF,Rel.Ab,MutID,stringsAsFactors = F)
+rm(VF,Rel.Ab,MutID)
+NoLigand=tab;rm(tab,file1)
+colnames(NoLigand)=paste("NoLig.",colnames(NoLigand),sep="")
+
+allMuts=union(NoLigand$NoLig.MutID,Library$Lib.MutID)
+idx=match(allMuts,Library$Lib.MutID)
+Mutation.Table=Library[idx,]
+idx=match(allMuts,NoLigand$NoLig.MutID)
+Mutation.Table=data.frame(allMuts,Mutation.Table,NoLigand[idx,],stringsAsFactors = F)
+rm(Library,NoLigand)
+
+# 
+# #--------------
+# allMuts=Mutation.Table$allMuts
+# searchI=gregexpr(":",allMuts,fixed = T)
+# start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+# chr=substr(allMuts,0,start.posI-1)
+# 
+# searchI=gregexpr(">",allMuts,fixed = T)
+# start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+# mut=substr(allMuts,start.posI-1,nchar(allMuts))
+# 
+# searchI=gregexpr(">",mut,fixed = T)
+# start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+# ref=substr(mut,start.posI-1,start.posI-1)
+# alt=substr(mut,start.posI+1,start.posI+1);rm(mut,searchI,start.posI)
+# 
+# searchI=gregexpr(":",allMuts,fixed = T)
+# start.pos=unlist(lapply(searchI, `[[`, 1),use.names = F)
+# searchI=gregexpr(">",allMuts,fixed = T)
+# end.pos=unlist(lapply(searchI, `[[`, 1),use.names = F)
+# start=end=substr(allMuts,start.pos+1,end.pos-2);rm(searchI,start.pos,end.pos)
+# 
+# AnnoDF=data.frame(chr,start,end,ref,alt,allMuts,stringsAsFactors = F);rm(chr,start,end,ref,alt)
+# 
+# # write.table(AnnoDF,file = "../20180410.InputAnnovar.AlliSCREAM.txt",col.names = F,quote = F,row.names = F)
+# # table_annovar.pl "20180410.InputAnnovar.Allvars.txt" ~/NGS_Seq_Tools/annovar2/humandb -buildver hg19 -out 20180410.EGFR.iSCREAM.AlliSCREAM -remove -protocol refgene,gnomad_genome,gnomad_exome,avsnp150,clinvar_20170905 -operation g,f,f,f,f -nastring . -csvout -polish
+
+
+
+#-------------
+# filtering by depth
+message("Starting with ",length(Mutation.Table$allMuts))
+filtered=Mutation.Table[Mutation.Table$Lib.DP>=1000&Mutation.Table$NoLig.DP>=1000,]
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"for DP>=1000. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+Mutation.Table=filtered
+
+
+# Add BamReadCount
+Lib.Count=readRDS("../../Bam-readcount/Parsed/parsed.EGFR.Lib.bamcount.RDS")
+Lib.Count$MutID=paste(Lib.Count$chr,":",Lib.Count$pos,Lib.Count$ref,">",Lib.Count$base,sep="")
+colnames(Lib.Count)=paste("bamRC.Lib.",colnames(Lib.Count),sep="")
+idx=match(Mutation.Table$allMuts,Lib.Count$bamRC.Lib.MutID)
+test=cbind.data.frame(Mutation.Table,Lib.Count[idx,])
+
+NoLig.Count=readRDS("../../Bam-readcount/Parsed/parsed.EGFR.NoLig.bamcount.RDS")
+NoLig.Count$MutID=paste(NoLig.Count$chr,":",NoLig.Count$pos,NoLig.Count$ref,">",NoLig.Count$base,sep="")
+colnames(NoLig.Count)=paste("bamRC.NoLig.",colnames(NoLig.Count),sep="")
+idx=match(Mutation.Table$allMuts,NoLig.Count$bamRC.NoLig.MutID)
+test=cbind.data.frame(test,NoLig.Count[idx,])
+
+Mutation.Table=test;rm(test,Lib.Count,NoLig.Count,idx)
+
+# Filtering based on strandBias
+filtered=Mutation.Table
+filtered=filtered[!((filtered$bamRC.NoLig.strandbias==Inf | filtered$bamRC.NoLig.strandbias==-Inf)|(filtered$bamRC.Lib.strandbias==Inf | filtered$bamRC.Lib.strandbias==-Inf)),]
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"for infinite SB. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+
+Mutation.Table=filtered
+filtered=filtered[filtered$bamRC.NoLig.strandbias<=10&filtered$bamRC.NoLig.strandbias>=-10|filtered$bamRC.Lib.strandbias<=10&filtered$bamRC.Lib.strandbias>=-10,]
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"for SB>10. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+
+Mutation.Table=filtered
+
+#--------------
+allMuts=Mutation.Table$allMuts
+searchI=gregexpr(":",allMuts,fixed = T)
+start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+chr=substr(allMuts,0,start.posI-1)
+
+searchI=gregexpr(">",allMuts,fixed = T)
+start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+mut=substr(allMuts,start.posI-1,nchar(allMuts))
+
+searchI=gregexpr(">",mut,fixed = T)
+start.posI=unlist(lapply(searchI, `[[`, 1),use.names = F)
+ref=substr(mut,start.posI-1,start.posI-1)
+alt=substr(mut,start.posI+1,start.posI+1);rm(mut,searchI,start.posI)
+
+searchI=gregexpr(":",allMuts,fixed = T)
+start.pos=unlist(lapply(searchI, `[[`, 1),use.names = F)
+searchI=gregexpr(">",allMuts,fixed = T)
+end.pos=unlist(lapply(searchI, `[[`, 1),use.names = F)
+start=end=substr(allMuts,start.pos+1,end.pos-2);rm(searchI,start.pos,end.pos)
+
+AnnoDF=data.frame(chr,start,end,ref,alt,allMuts,stringsAsFactors = F);rm(chr,start,end,ref,alt)
+
+# write.table(AnnoDF,file = "../20180401.InputAnnovar.Allvars.txt",col.names = F,quote = F,row.names = F)
+# table_annovar.pl "20180401.InputAnnovar.Allvars.txt" ~/NGS_Seq_Tools/annovar2/humandb -buildver hg19 -out 20180401.EGFR.iSCREAM.Allvars -remove -protocol refgene,gnomad_genome,gnomad_exome,avsnp150,clinvar_20170905 -operation g,f,f,f,f -nastring . -csvout -polish
+
+
+rm(AnnoDF)
+var=read.table("../20180401.EGFR.iSCREAM.Allvars.hg19_multianno.csv",sep=",",stringsAsFactors = F,header = T)
+var$MutID=paste(var$Chr,":",var$Start,var$Ref,">",var$Alt,sep="")
+
+idx=match(Mutation.Table$allMuts,var$MutID)
+Mutation.Table=data.frame(Mutation.Table,var[idx,c(6:33)]);rm(var)
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/EleniusGroup/DvsP.EGFR.HiDP/AnnovarMutCodeFind.R")
+
+Mutation.Table$AAchange=annovarMutCodeFind(Mutation.Table$AAChange.refgene,isoform = "NM_005228")
+temp=gsub("^.","",x = Mutation.Table$AAchange);temp=gsub(".$","",x=temp)
+Mutation.Table$AAPos=as.numeric(temp);rm(temp)
+
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/EleniusGroup/DvsP.EGFR.HiDP/Annovar_cDNA_Find.R")
+Mutation.Table$cDNAchange=annovar_cDNA_Find(Mutation.Table$AAChange.refgene,isoform = "NM_005228")
+temp=gsub("^.","",x = Mutation.Table$cDNAchange);temp=gsub(".$","",x=temp)
+Mutation.Table$cDNAPos=as.numeric(temp);rm(temp)
+
+
+
+# keeping exonic
+filtered=Mutation.Table[Mutation.Table$Func.refgene=="exonic",] # because only looking at cDNA variants so exonic
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"for exonic. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+Mutation.Table=filtered
+# Keeping common mutations
+
+filtered=droplevels.data.frame(na.exclude(Mutation.Table))
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"Uncommon between samples. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+Mutation.Table=filtered
+
+
+# temp=droplevels.data.frame(na.exclude(Mutation.Table))
+
+# removing synonymous
+Mutation.Table=filtered
+filtered=Mutation.Table[Mutation.Table$ExonicFunc.refgene!="synonymous SNV",]
+message(paste("Filtered out",(length(Mutation.Table$allMuts)-length(filtered$allMuts)),"synonymous variants. Remaining =",length(filtered$allMuts),"unique AAs=",length(unique(filtered$AAchange))))
+Mutation.Table=filtered
+rm(filtered)
+
+
+# saveRDS(object =Mutation.Table,file = "../20180401.MutationTable.RDS")
+# write.table(x = Mutation.Table,file = "../20180401.MutationTable.tsv",sep="\t",row.names = F,col.names = T,quote = F)
+
+
+
+load("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/EleniusGroup/ERBB.Miscellaneous/Posssible mutants/EGFR/egfr_Variants.bin")
+print(paste(length(unique(Variants$mutants)),length(unique(Variants$mutantprotein))))
+
+

5.Depth&SummaryStats.R

+library(ggplot2)
+#----------------------
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf")
+
+rm(list=ls())
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+Mutation.Table=readRDS("20180401.MutationTable.RDS")
+Mutation.Table$FC.reg=(Mutation.Table$NoLig.VF/Mutation.Table$Lib.VF)
+
+# source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/SignedCalcs/SignedFoldChange.R")
+# Mutation.Table$FC=FoldChange(Mutation.Table$Lib.VF,Mutation.Table$NoLig.VF)
+# saveRDS(object = Mutation.Table,file = "20180411.MutationTable.iSCREAM.RDS")
+
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Misc Code/Master ggplot2 Theme/theme.DC.plot.R")
+
+#----------------------------------
+subdf=Mutation.Table[,c("allMuts","Lib.DP","NoLig.DP")]
+subdf.long=reshape2::melt(data = subdf)
+
+#ggplot(data=subdf.long,aes(x=value,group=variable))+geom_histogram(color="black",fill="cornflowerblue",binwidth = 2000)+facet_wrap(~variable)+theme.drugs.iscream+scale_x_continuous(breaks = seq(0,max(subdf.long$value),by=50000),labels=scales::comma)
+
+y.breaks=seq(0,ceiling(max(subdf.long$value,na.rm = T)),by = 50000)
+ggplot(data = subdf.long,aes(x=variable,y=value))+geom_jitter(height = 0,alpha=0.3,size=1,color="grey70")+geom_boxplot(outlier.color = NA,color="black",fill=NA)+scale_y_continuous(breaks = y.breaks,limits = c(0,max(subdf.long$value,na.rm = T)),labels = scales::comma)+ylab("Total Depth for each variant")+xlab("Sample")+theme.drugs.iscream
+# ggsave(filename = "20180401.DP.box.pdf",width = 3,height = 4)
+
+ggplot(data = subdf.long,aes(x=variable,y=value))+geom_jitter(height = 0,alpha=0.3,size=1,color="grey70")+geom_violin(color="black",fill=NA,trim = F,adjust=2,na.rm = T,scale = "width",width=0.8)+geom_boxplot(outlier.color = NA,color="black",fill=NA,width=0.1)+scale_y_continuous(breaks = y.breaks,limits = c(0,max(subdf.long$value,na.rm = T)),labels = scales::comma)+ylab("Total Depth for each variant")+xlab("Sample")+theme.drugs.iscream
+# ggsave(filename = "20180401.DP.violin.pdf",width = 3,height = 4)
+
+
+subdf=Mutation.Table[,c("allMuts","Lib.AD","NoLig.AD")]
+subdf.long=reshape2::melt(data = subdf)
+# subdf.long$value=subdf.long$value+1
+
+base=c(1,2,4,6,8)
+y.breaks=c(base*1,base*10,base*100,base*1000,10000)
+ggplot(data = subdf.long,aes(x=variable,y=log10(value)))+geom_jitter(height = 0,alpha=0.3,size=1,color="grey70")+geom_boxplot(outlier.color = NA,color="black",fill=NA)+scale_y_continuous(breaks = log10(y.breaks),labels = c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000),limits = c(0,4.1))+ylab("Reads supporting each variant")+xlab("Sample")+theme.drugs.iscream
+# # ggsave(filename = "20180401.AD.box.pdf",width = 3,height = 4)
+
+ggplot(data = subdf.long,aes(x=variable,y=log10(value)))+geom_jitter(height = 0,alpha=0.3,size=1,color="grey70")+geom_violin(color="black",fill=NA,trim = F,adjust=2,na.rm = T,scale = "width",width=0.8)+geom_boxplot(outlier.color = NA,color="black",fill=NA,width=0.1)+scale_y_continuous(breaks = log10(y.breaks),labels = c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000),limits = c(0,4.1))+ylab("Reads supporting each variant")+xlab("Sample")+theme.drugs.iscream
+# ggsave(filename = "20180401.AD.violin.pdf",width = 3,height = 4)
+
+#-------------------------
+# Fold change
+rm(subdf.long,subdf)
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/SignedCalcs/SignedFoldChange.R")
+subdf=Mutation.Table[,c("allMuts","AAchange","Lib.VF","NoLig.VF","FC.reg"),]
+
+subdf$FC=FoldChange(subdf$Lib.VF,subdf$NoLig.VF)
+
+base=5;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 6912
+base=10;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 7068
+base=15;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 7123
+base=20;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 7138
+base=25;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 7149
+base=30;(dim(subdf[subdf$FC>=(-1*base)&subdf$FC<=base,])) # 7164
+
+base=5;(dim(subdf[subdf$FC>=base,])) # 120
+base=10;(dim(subdf[subdf$FC>=base,])) # 40
+base=15;(dim(subdf[subdf$FC>=base,])) # 28
+base=20;(dim(subdf[subdf$FC>=base,])) # 23
+base=25;(dim(subdf[subdf$FC>=base,])) # 21
+base=30;(dim(subdf[subdf$FC>=base,])) # 15
+
+base=0.5;(dim(subdf[subdf$NoLig.VF>=base,])) # 33
+base=1;(dim(subdf[subdf$NoLig.VF>=base,])) # 12
+base=2;(dim(subdf[subdf$NoLig.VF>=base,])) # 6
+base=3;(dim(subdf[subdf$NoLig.VF>=base,])) # 5
+base=4;(dim(subdf[subdf$NoLig.VF>=base,])) # 1
+base=5;(dim(subdf[subdf$NoLig.VF>=base,])) # 0
+
+base=0.5;(dim(subdf[subdf$Lib.VF>=base,])) # 192
+base=1;(dim(subdf[subdf$Lib.VF>=base,])) # 114
+base=2;(dim(subdf[subdf$Lib.VF>=base,])) # 39
+base=3;(dim(subdf[subdf$Lib.VF>=base,])) # 13
+base=4;(dim(subdf[subdf$Lib.VF>=base,])) # 5
+base=5;(dim(subdf[subdf$Lib.VF>=base,])) # 1
+base=6;(dim(subdf[subdf$Lib.VF>=base,])) # 1
+base=7;(dim(subdf[subdf$Lib.VF>=base,])) # 0
+
+
+# ggplot(data = subdf,aes(x=FC.reg))+stat_bin(breaks=c(0,1,2.5,5,7.5,10,15,25,50,75,100,150),color='black',fill="grey50")+stat_bin(breaks=c(0,1,2.5,5,7.5,10,15,25,50,75,100,150), geom="text", aes(label=..count..), vjust=-0.5,size=2)+theme.drugs.iscream+ylab("No. of observations")+xlab("FoldChange")+scale_x_continuous(breaks = seq(0,150,by = 25))+ggtitle("Distribution of Fold Change")
+
+
+ggplot(data = subdf,aes(x=FC.reg))+stat_bin(binwidth=5,color='black',fill="grey50")+stat_bin(binwidth=5, geom="text", aes(label=..count..), vjust=-0.5,size=2)+theme.drugs.iscream+ylab("No. of observations")+xlab("FoldChange")+scale_x_continuous(breaks = seq(0,150,by = 25))+ggtitle("Distribution of Fold Change")
+ggsave(filename = "20180410.FC.reg.Hist.bw5.pdf",width = 6,height = 3)
+
+
+
+# base=c(1,2,4,6,8)
+# y.breaks=c(base*1,base*10,base*100,base*1000,10000)
+# ggplot(data = subdf,aes(x=FC))+stat_bin(binwidth=5,color='black',fill="grey50")+stat_bin(binwidth=5, geom="text", aes(label=..count..), vjust=-0.5,size=2)+theme.drugs.iscream+scale_y_log10(breaks=c(y.breaks),labels=c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000))+ylab("No. of observations (log10 scale)")+xlab("FoldChange")+scale_x_continuous(breaks = seq(-250,150,by = 25))+ggtitle("Distribution of Fold Change")
+# # ggsave(filename = "20180401.FC.Log.Hist.pdf",width = 10,height = 6)
+#  
+# ggplot(data = subdf,aes(x=FC))+stat_bin(binwidth=5,color='black',fill="grey50")+stat_bin(binwidth=5, geom="text", aes(label=..count..), vjust=-0.5,size=2)+theme.drugs.iscream+ylab("No. of observations")+ggtitle("Distribution of Fold Change")+scale_y_continuous(breaks = seq(0,8000,by = 500))+scale_x_continuous(breaks = seq(-250,150,by = 25))
+# # ggsave(filename = "20180401.FC.Hist.pdf",width = 10,height = 6)
+# 
+# ggplot(data = subdf,aes(x=FC))+stat_bin(binwidth=5,color='black',fill="grey50")+stat_bin(binwidth=5, geom="text", aes(label=..count..), vjust=-0.5,size=2)+theme.drugs.iscream+ylab("No. of observations")+xlab("FoldChange")+scale_x_continuous(breaks = seq(-250,150,by = 25))+ggtitle("Distribution of Fold Change")+scale_y_continuous(breaks = seq(0,8000,by = 500))
+# # ggsave(filename = "20180401.FC.Hist.tall.pdf",width = 10,height = 20)
+bw=0.1
+ggplot(data = subdf,aes(x=NoLig.VF))+geom_histogram(binwidth=bw,color='black',fill="grey50")+stat_bin(binwidth=bw, geom="text", aes(label=..count..),hjust=-0.2,vjust=-0.2,size=2,angle = 45)+theme.drugs.iscream+ylab(paste("No. of observations"))+xlab("Variant Frequency (%)")+scale_x_continuous(breaks = seq(0,7,by = 0.5))+ggtitle("Distribution of Variant Frequency (NoLigand)")+scale_y_continuous(breaks = seq(0,6000,by = 1000),limits = c(0,6000))
+ggsave(filename = "20180410.VF.NoLig.pdf",width = 6,height = 3)
+
+# bw=0.1
+# p1=ggplot(data = subdf,aes(x=Lib.VF))+geom_histogram(binwidth=bw,color='black',fill="grey50")+stat_bin(binwidth=bw, geom="text", aes(label=..count..),hjust=-0.2,vjust=-0.2,size=2,angle = 45)+theme.drugs.iscream+ylab(paste("No. of observations"))+xlab("Variant Frequency (%)")+scale_x_continuous(breaks = seq(0,7,by = 0.5),limits = c(-0.1,6.2))+ggtitle("Distribution of Variant Frequency (Library)")+scale_y_continuous(breaks = seq(0,6000,by = 1000),limits = c(0,6000));p1
+# p2=ggplot(data = subdf,aes(x=NoLig.VF))+geom_histogram(binwidth=bw,color='black',fill="grey50")+stat_bin(binwidth=bw, geom="text", aes(label=..count..),hjust=-0.2,vjust=-0.2,size=2,angle = 45)+theme.drugs.iscream+ylab(paste("No. of observations"))+xlab("Variant Frequency (%)")+scale_x_continuous(breaks = seq(0,7,by = 0.5),limits = c(-0.1,6.2))+ggtitle("Distribution of Variant Frequency (NoLigand)")+scale_y_continuous(breaks = seq(0,6000,by = 1000),limits = c(0,6000));p2
+# 
+# gp=gridExtra::grid.arrange(p1, p2)
+# ggsave(plot = gp,filename = "20180401.VF.Hist.pdf",width = 10,height = 6)
+
+
+# subdf.long=reshape2::melt(data = subdf[,c("allMuts","Lib.VF","NoLig.VF")])
+# ggplot(data=subdf.long,aes(x=variable,y=value))+geom_jitter(height = 0,alpha=0.3,size=1,color="grey70")+geom_boxplot(outlier.color = NA,color="black",fill=NA)+theme.drugs.iscream
+
+#------------------------------
+
+# dups=duplicated(Mutation.Table$AAchange)
+
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/SignedCalcs/SignedFoldChange.R")
+Mutation.Table$FC=FoldChange(Mutation.Table$Lib.VF,Mutation.Table$NoLig.VF)
+
+subdf=Mutation.Table[,c("FC","AAchange","AAPos","allMuts","Lib.VF","NoLig.VF")]
+
+dups=subdf$AAchange[duplicated(subdf$AAchange)];length(unique(dups))
+# match(x = dups,table = subdf$AAchange)
+# which(dups %in% subdf$AAchange)
+dups.df=subdf[which(subdf$AAchange %in% dups),]
+
+length(unique(dups.df$AAchange))
+
+ggplot(data=dups.df,aes(x=AAPos,y=FC,group=AAchange))+geom_point()+geom_line()+geom_text(data=dups.df[dups.df$FC>15,],aes(label=dups.df[dups.df$FC>15,"AAchange"]),hjust=0.5,vjust=-0.8,size=3)+geom_text(data=dups.df[dups.df$FC<(-15),],aes(label=dups.df[dups.df$FC<(-15),"AAchange"]),hjust=0.5,vjust=1.8,size=3)+theme.drugs.iscream+ggtitle(paste("FC Dist for Degenerate Mutations. Total dots =",length(dups.df$AAchange)))+ylab("Fold Change")+xlab("EGFR AA position")+geom_rug(sides="r",alpha=0.5,size=0.2)
+# ggsave(filename = "20180401.DegenrateMuts.FC.pdf",width = 10,height = 7)
+
+
+#cols=colorRampPalette(c("#003366","firebrick1"))(20)
+
+ggplot(data=dups.df,aes(x=AAPos,y=FC,group=AAchange))+geom_point(alpha=0.5,aes(color=dups.df$NoLig.VF,size=dups.df$NoLig.VF))+geom_point(alpha=0.5,aes(color=dups.df$Lib.VF,size=dups.df$Lib.VF))+geom_line()+geom_text(data=dups.df[dups.df$FC>15,],aes(label=dups.df[dups.df$FC>15,"AAchange"]),hjust=0.5,vjust=-0.8,size=3)+geom_text(data=dups.df[dups.df$FC<(-15),],aes(label=dups.df[dups.df$FC<(-15),"AAchange"]),hjust=0.5,vjust=1.8,size=3)+ggtitle(paste("FC Dist for Degenerate Mutations. Total dots =",length(dups.df$AAchange)))+ylab("Fold Change")+xlab("EGFR AA position")+scale_colour_gradient(low="#003366",high="firebrick1")+theme.drugs.iscream.legend
+# ggsave(filename = "20180401.DegenrateMuts.FC.with.VF.pdf",width = 11,height = 7)
+
+#------------------------
+#----------------------
+setwd("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf")
+
+rm(list=ls())
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+Mutation.Table=readRDS("20180401.MutationTable.RDS")
+
+source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Misc Code/Master ggplot2 Theme/theme.DC.plot.R")
+
+#----------------------------------
+subdf=Mutation.Table[,c("allMuts","Lib.DP","NoLig.DP","AAchange","AAPos","cDNAchange")]
+df=subdf[,c(6,2,3)]
+subdf.long=reshape2::melt(df)
+temp=gsub("^.","",x =subdf.long$cDNAchange);temp=gsub(".$","",x=temp)
+subdf.long$cDNAPos=as.numeric(temp);rm(temp)
+
+apply
+
+ggplot(data=subdf.long,aes(x=cDNAPos,y=value))+geom_point(size=0.2,aes(color=variable,alpha=0.2))+geom_line()+theme_dc.plot2+scale_x_continuous(breaks = c(0,seq(1,3500,by = 200),3633))+scale_y_continuous(breaks =c(seq(0,300000,by = 50000)))+scale_color_manual(values=c("black","red"))
+
+
+# base=c(1,2,4,6,8)
+# y.breaks=c(base*1,base*10,base*100,base*1000,base*10000,base*100000)
+# ggplot(data=subdf.long,aes(color=variable,x=cDNAPos,y=log10(value)))+geom_point(size=0.2)+geom_line()+theme_dc.plot2+scale_x_continuous(breaks = c(0,seq(1,3500,by = 200),3633))+scale_y_continuous(breaks = log10(y.breaks),limits = c(0,5.5),labels =c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000,rep("",4),100000,rep("",4)))

AnnovarMutCodeFind.R

+#MutInfo# Function Description
+# This function takes a mutation info column from ANNOVAR output and selects the protein change based on the isoform provided.
+# Isoform accepted as GI identifier in the block for the particular isoform in the ANNOVAR mutation info. You should look at the data and then provide the info. Defaults to ERBB4 A2 isoform.
+
+annovarMutCodeFind=function(MutationColumn,isoform="NM_005228"){
+  MutationList=c("List of mutations")
+  for(i in seq(1:length(MutationColumn))){
+    MutInfo=MutationColumn[i]
+    l=sort(unique(unlist(strsplit(MutInfo,","))))
+    l2=l[grep(isoform,l)]
+    l2.s=unique(unlist(strsplit(l2,":")))
+    l3=l2.s[grep("^p",l2.s)]
+    l3=gsub("p.","",l3)
+    l3=gsub("X","*",l3)
+    MUTATION=l3
+    if(length(MUTATION)==0) MUTATION=" "
+    MutationList=c(MutationList,MUTATION)
+  }
+  return(MutationList[-1])
+}
+
+
+# MutationColumn=Mutation.Table$AAChange.refgene;i=7591;isoform="NM_005228"
+# annovarMutCodeFind(tab.s$Mutation)

Annovar_cDNA_Find.R

+#MutInfo# Function Description
+# This function takes a mutation info column from ANNOVAR output and selects the cDNA change based on the isoform provided.
+# Isoform accepted as GI identifier in the block for the particular isoform in the ANNOVAR mutation info. You should look at the data and then provide the info. Defaults to ERBB4 A2 isoform.
+
+annovar_cDNA_Find=function(MutationColumn,isoform="NM_005228"){
+  MutationList=c("List of mutations")
+  for(i in seq(1:length(MutationColumn))){
+    MutInfo=MutationColumn[i]
+    l=sort(unique(unlist(strsplit(MutInfo,","))))
+    l2=l[grep(isoform,l)]
+    l2.s=unique(unlist(strsplit(l2,":")))
+    l3=l2.s[grep("^c",l2.s)]
+    l3=gsub("c.","",l3)
+    #l3=gsub("X","*",l3)
+    MUTATION=l3
+    if(length(MUTATION)==0) MUTATION=" "
+    MutationList=c(MutationList,MUTATION)
+  }
+  return(MutationList[-1])
+}
+
+
+# MutationColumn=Mutation.Table$AAChange.refgene;i=7591;isoform="NM_005228"
+# annovarMutCodeFind(tab.s$Mutation)

COSMIC figure.R

+library(ggplot2)
+rm(list=ls())
+setwd("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Klaus Lab/Manuscripts/EGFR Screen Manuscript/Data/Supplementary figures/COSMIC/")
+source("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/MutSiteFind/MutSiteFind.R")
+source("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Misc Code/Master ggplot2 Theme/theme.DC.plot.R")
+
+
+data=readr::read_csv("V84_38_MUTANT.csv",col_names = T)
+data=data.frame(data)
+data=na.exclude(data)
+data$SAMPLE_NAME=toupper(data$SAMPLE_NAME)
+
+data$MUTATION_AA=gsub("^p.","",data$MUTATION_AA)
+data$MutID=paste(data$SAMPLE_NAME,"=",data$MUTATION_AA,sep="")
+data.bak=data
+data=data[-which(duplicated(data$MutID)),]
+data=data[-grep("\\?",data$MUTATION_AA),]
+
+#data=data[data$Mutation.Type!="Fusion",]
+
+uniqueMuts=table(data$MUTATION_AA)
+df=data.frame(AAchange=names(uniqueMuts),count=uniqueMuts,stringsAsFactors = F)
+df=df[,-2];colnames(df)=c("AAchange","count")
+df$AAPos=as.numeric(MutSiteFind(df$AAchange))
+df$Type=data$MUTATION_DESCRIPTION[match(df$AAchange,data$MUTATION_AA)]
+df=na.exclude(df)
+df=subset(df,subset = df$Type!="Substitution - coding silent")
+
+# temp=df[(df$AAPos%in%c(746,747)),]
+temp=df[(df$AAPos%in%seq(746,752)),]
+temp=temp[temp$Type!="Substitution - Missense",]
+temp=temp[-grep("Frameshift",temp$Type),]
+df=df[-match(x = temp$AAchange,table = df$AAchange),]
+df=rbind(df,c("exon19del",sum(temp$count),746,"Complex - deletion inframe"))
+df$count=as.numeric(df$count)
+df$AAPos=as.numeric(df$AAPos)
+
+rang=c("brown","brown","black","brown","black","brown","black","black","brown","#198c19","red")
+names(rang)=sort(unique(df$Type))
+
+base=c(1,2,4,6,8)
+breaks=c(base*1,base*10,base*100,base*1000,10000)
+labels=c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000)
+ggplot(df,aes(x = AAPos,y=log10(count)))+geom_segment(aes(yend=-0.4,xend=df$AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+xlab("EGFR Amino Acid Position")+ylab("No. of mutations (log10 scale)")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")+scale_y_continuous(breaks = log10(breaks),labels=labels,limits = c(-0.4,4.1))+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations COSMIC, N=",length(data$MutID)))
+
+# ggsave("log.20180330 EGFR_lollipop COSMIC.pdf",width = 10,height = 2.5)
+
+
+ggplot(df,aes(x = AAPos,y=count))+geom_segment(aes(yend=-20,xend=df$AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+xlab("EGFR Amino Acid Position")+ylab("No. of mutations")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")+scale_color_manual(values=rang)
+
+# ggsave("reg.20180330 EGFR_lollipop COSMIC.pdf",width = 10,height = 2.5)
+
+
+
+base=c(1,2,4,6,8)
+breaks=c(base*1,base*10,base*100,base*1000,10000)
+labels=c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000)
+ggplot(df,aes(x = AAPos,y=log10(count)))+geom_segment(aes(yend=-0.4,xend=df$AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+xlab("EGFR Amino Acid Position")+ylab("No. of mutations (log10 scale)")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")+scale_y_continuous(breaks = log10(breaks),labels=labels,limits = c(-0.4,4.1))+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations COSMIC, N=",length(data$MutID)))+geom_rug(sides="t",alpha=0.5,color="black",size=0.2)
+
+# ggsave("log.20180330 EGFR_lollipop COSMIC.rug.pdf",width = 10,height = 2.5)
+
+
+#---------------
+# only hits from screen
+
+# source("~/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto//Git/GitLab.UTU/Personal/str.extra-for-R.R/str.extra.R")
+Mutation.Table=readRDS("~/BaseSpace/20180112 EGFR Trimmed HiDP/vcf/20180411.MutationTable.iSCREAM.RDS")
+sub=Mutation.Table[Mutation.Table$FC.reg>=25,];rm(Mutation.Table)
+
+#---------------------
+subdf=sub[,c("AAchange","AAPos","cDNAchange","cDNAPos","FC.reg","allMuts")]
+colnames(subdf)=paste("sdf.",colnames(subdf),sep="")
+
+idx=match(x = subdf$sdf.AAchange,table = df$AAchange)
+subdf=cbind(subdf,df[idx,])
+
+subdf$count[is.na(subdf$count)]=0.1
+subdf$Type[is.na(subdf$Type)]="absent"
+subdf$Type[is.na(subdf$Type)]="absent"
+subdf$AAPos[is.na(subdf$AAPos)]=0
+subdf$AAchange[is.na(subdf$AAchange)]=""
+
+
+rang=c("gray","#198c19","red")
+names(rang)=sort(unique(subdf$Type))
+
+base=c(1,2,4,6,8)
+breaks=c(base*1,base*10,base*100,base*1000,10000)
+labels=c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000)
+ggplot(subdf,aes(x = sdf.AAPos,y=log10(count)))+geom_segment(aes(yend=-1,xend=subdf$sdf.AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+xlab("EGFR Amino Acid Position")+ylab("No. of mutations (log10 scale)")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+scale_y_continuous(breaks = log10(breaks),labels=labels,limits = c(-1,4.2))+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations COSMIC, N=",length(data$MutID)))+geom_text(data=subdf,aes(label=subdf$sdf.AAchange),vjust=-0.3,size=2)
+# +geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")
+ggsave("log.20180411 EGFR_iSCREAM COSMIC same change.pdf",width = 10,height = 2.5)
+rm(idx,subdf,rang,base,breaks)
+#---------------------
+
+source("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/MutSiteFind/sameSiteChangesFind.R")
+
+uniquePos=unique(df$AAPos)
+change=c()
+count=c()
+for(pos in uniquePos){
+  subdf=df[df$AAPos==pos,]
+  count=c(count,sum(subdf$count))
+  subdf=subdf[order(subdf$count,decreasing = T),]
+  temp=sameSiteChangesFind(subdf$AAchange)
+  change=c(change,paste(temp,collapse="/"));rm(temp)
+}
+
+newdf=data.frame(uniquePos,change,count,stringsAsFactors = F)
+
+#---------------
+subdf=sub[,c("AAchange","AAPos","cDNAchange","cDNAPos","FC.reg","allMuts")]
+colnames(subdf)=paste("sdf.",colnames(subdf),sep="")
+
+idx=match(x = subdf$sdf.AAPos,table = newdf$uniquePos)
+subdf=cbind(subdf,newdf[idx,])
+sum(subdf$count,na.rm = T)
+
+subdf$count[is.na(subdf$count)]=0.1
+subdf$uniquePos[is.na(subdf$uniquePos)]=0
+subdf$change[is.na(subdf$change)]=""
+subdf$Type=rep("",length(subdf$sdf.AAchange))
+subdf$Type[subdf$change!=""]="Missense"
+subdf$Type[subdf$Type!="Missense"]="absent"
+subdf$Type[subdf$change=="*"]="Nonsense"
+
+subdf$label=paste(substr(subdf$sdf.AAchange,start=0,stop=nchar(subdf$sdf.AAchange)-1),subdf$change,sep="")
+
+subdf$label
+
+rang=c("gray","#198c19","red")
+names(rang)=sort(unique(subdf$Type))
+
+base=c(1,2,4,6,8)
+breaks=c(base*1,base*10,base*100,base*1000,10000)
+labels=c(1,rep("",4),10,rep("",4),100,rep("",4),1000,rep("",4),10000)
+ggplot(subdf,aes(x = sdf.AAPos,y=log10(count)))+geom_segment(aes(yend=-1,xend=subdf$sdf.AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+xlab("EGFR Amino Acid Position")+ylab("No. of mutations (log10 scale)")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+scale_y_continuous(breaks = log10(breaks),labels=labels,limits = c(-1,4.2))+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations COSMIC, N=",length(data$MutID)))+geom_text(data=subdf,aes(label=subdf$label),vjust=-0.3,size=2)
+
+ggsave("log.20180411 EGFR_iSCREAM COSMIC same residue.pdf",width = 10,height = 2.5)
+
+sum(subdf$count)
+

cBioportal Figure.R

+library(ggplot2)
+rm(list=ls())
+setwd("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Klaus Lab/Manuscripts/EGFR Screen Manuscript/Data & Figures/Supplementary figures/cBioportal/")
+source("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/Gitlab.DC/MutSiteFind/MutSiteFind.R")
+source("/Users/Deepankar/Documents/OneDrive.UTU/OneDrive - O365 Turun yliopisto/Git/GitLab.UTU/Misc Code/Master ggplot2 Theme/theme.DC.plot.R")
+
+
+data=readr::read_tsv("20180321 cBioportal EGFR mutations unique samples.tsv",col_names = T)
+data=data.frame(data)
+data$Sample.ID=toupper(data$Sample.ID)
+data$MutID=paste(data$Sample.ID,"=",data$Protein.Change,sep="")
+
+
+data.bak=data
+data=data[-which(duplicated(data$MutID)),]
+data=data[data$Mutation.Type!="Fusion",]
+
+uniqueMuts=table(data$Protein.Change)
+df=data.frame(AAchange=names(uniqueMuts),count=uniqueMuts,stringsAsFactors = F)
+df=df[,-2];colnames(df)=c("AAchange","count")
+df$Type=data$Mutation.Type[match(df$AAchange,data$Protein.Change)]
+df$AAPos=as.numeric(MutSiteFind(df$AAchange))
+df=na.exclude(df)
+# df=subset(df,subset = df$Type!="Fusion")
+
+rang=c("#198c19","brown","black","brown","#198c19","black","black","brown","brown","black","black")
+names(rang)=unique(df$Type)
+
+ggplot(df,aes(x = AAPos,y=count))+geom_segment(aes(yend=-20,xend=df$AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+scale_y_continuous(breaks = c(seq(0,max(df$count),by = 50),max(df$count)),limits = c(-30,max(df$count)))+xlab("EGFR Amino Acid Position")+ylab("No. of mutations")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations cBioPortal, N=",length(data$MutID)))
+
+ggsave("20180330 EGFR_lollipop cBioportal.pdf",width = 10,height = 2.5)
+
+
+ggplot(df,aes(x = AAPos,y=count))+geom_segment(aes(yend=-20,xend=df$AAPos))+geom_point(size=1.5,alpha=0.65,aes(color=Type))+theme_dc.plot2+scale_y_continuous(breaks = c(seq(0,max(df$count),by = 50),max(df$count)),limits = c(-30,max(df$count)))+xlab("EGFR Amino Acid Position")+ylab("No. of mutations")+scale_x_continuous(breaks = c(1,seq(100,1200,by = 100),1210))+geom_vline(xintercept = c(1,24,75,300,390,600,646,668,712,979),color="black",linetype="dotted")+scale_color_manual(values=rang)+ggtitle(paste("EGFR mutations cBioPortal, N=",length(data$MutID)))+geom_rug(sides="t",alpha=0.5,color="black",size=0.2)
+
+ggsave("20180330 EGFR_lollipop cBioportal.rug.pdf",width = 10,height = 2.5)
+
+# ggplot(FC.data.cut,aes(x=FoldChange,y=reorder(Mutation,-Lib.AAPos)))+geom_segment(aes(yend=FC.data.cut$Mutation),xend=0,color="black")+geom_text(data=FC.data.cut[FC.data.cut$FoldChange>8,],aes(label=FC.data.cut[FC.data.cut$FoldChange>8,"Mutation"],hjust=-0.3),size=0.8)+geom_text(data=FC.data.cut[FC.data.cut$FoldChange<(-8),],aes(label=FC.data.cut[FC.data.cut$FoldChange<(-8),"Mutation"],hjust=1.2),size=0.8)+geom_point(size=0.5,aes(color=Direction))+scale_color_manual(values=c("Blue","Red"))+geom_vline(xintercept = 0)+geom_vline(xintercept = 10,linetype="dashed")+geom_vline(xintercept = -10,linetype="dashed")+theme(panel.border = element_rect(linetype = 1, colour = "black",fill=NA,size=0.15),panel.background=element_rect(fill = NA, colour = NA),axis.text= element_text(size = rel(0.2),color="black"),legend.key= element_rect(fill=NA,colour = NA), axis.ticks =element_line(colour = "black"))+ggtitle("FC Library vs No Ligand (All Mutations)")+scale_x_continuous(breaks = seq(-50,34,by =2),limits = c(-50,34))+xlab("Fold Change")+ylab("Mutations")
+
+
+data=data.bak
+temp=data[data$Protein.Change=="T790M",]
+temp=temp[-which(duplicated(temp$MutID)),]
+
+temp2=data[,]
+
+length(unique(data$MutID))
+
+sub=subset(data,duplicated(data$MutID))
+