##Lets try and do what I was doing in excel for TE's in R
TEdata<-read.csv("C:\\Users\\Harry\\Documents\\coordinates4R.csv")
TEdata<-TEdata[5:nrow(TEdata),]

#s1=TEdata$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1
#e1=TEdata$X
#s2=TEdata$X.2
#e2=TEdata$X.3
#t1=TEdata$X.10
#t2=TEdata$X.11
nrow(TEdata)
virginData=TEdata
TEdata<-subset(TEdata, as.numeric(levels(X)[X])!=as.numeric(levels(X.3)[X.3]) | levels(X.10)[X.10]!=levels(X.11)[X.11] | as.numeric(levels(X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1)[X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1])!=as.numeric(levels(X.2)[X.2]))
nrow(TEdata)
#can we make TEdata smaller
# newDat=rep(-1,16)
a=c(0)
selfsRemoved=TEdata
done=FALSE

id<-1:nrow(TEdata)
TEdata<-cbind(TEdata,id)
goodData=c(-1)
badData=c(-1)
#The easiest way of fixing the problems I was having is to make a newdata set from the old one
for(i in id){
  #Check row i
  currentRow=TEdata[i,]
  start1=as.numeric(levels(currentRow$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1)[currentRow$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1])
  end1=as.numeric(levels(currentRow$X)[currentRow$X])
  start2=as.numeric(levels(currentRow$X.2)[currentRow$X.2])
  end2=as.numeric(levels(currentRow$X.3)[currentRow$X.3])
  tag1=currentRow$X.10
  tag2=currentRow$X.11
  A=levels(tag1)[tag1]
  #Is there another row that matches it?
  otherRow=subset(TEdata, X.2==start1 & X.3==end1 & X.11==A & X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1==start2 & X==end2 & X.10==levels(tag2)[tag2])
  cID=otherRow$id
  C=currentRow$id
  skip=FALSE
  L=nrow(otherRow)
  if(L==0){
    skip=TRUE
  }
  if(!skip & !C%in%badData){#IF there is
    goodData=rbind(goodData,C)
    badData=rbind(badData,cID)
  }else if(skip){
    goodData=rbind(goodData,C)
  }
}
goodData=subset(goodData, goodData>0)
badData=subset(badData, badData>0)
nrow(TEdata)

TEdata=subset(TEdata, id%in%goodData)
write.csv(TEdata, file = "halfsize_coordinates_file.csv")
# #At this point all of the data points that are exactly the same have been removed
# #now we want to remove by some criterion (subject sequence matched >10X)
# 


#cleanData=cbind(TEdata$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1, TEdata$X,TEdata$X.2,TEdata$X.3,TEdata$X.5, TEdata$X.6, TEdata$X.8, TEdata$X.10, TEdata$X.11, TEdata$id)
# #s1=TEdata$X.10
# data=rep(-1,16)

#First we will look at a particular gene than we need to just look at one location
#so we should need three nested loops
# criticalN=10
# for(tag in s1){
#   currentGene=subset(TEdata,X.10==tag)
#   startPositions=levels(currentGene$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1)
#   if(length(currentGene$X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1)>=criticalN){
#     for(start in startPositions){
#       currentStart=subset(currentGene,X.home.harry.Documents.Bioinformatics.cgigas_alpha_v032.fa.1==start)
#       endPositions=levels(currentStart$X)
#       if(length(currentStart$X)>=criticalN){
#         for(end in endPositions){
#           currentLocation=subset(currentStart, X==end)
#           repeats=nrow(currentLocation)
#           if(repeats>=criticalN){
#             data=rbind(data,currentLocation)
#           }
#         }
#       }
#     }
#   }
# }
# x=2
# x