comparison report_clonality/RScript.r @ 43:2325074a8461 draft

Uploaded

42:9a47d7a552d6

# ---------------------- AA median CDR3 length ----------------------

median.aa.l = data.frame(data.table(PRODF)[, list(median=as.double(median(as.numeric(.SD$CDR3.Length, na.rm=T), na.rm=T))), by=c("Sample")])
write.table(median.aa.l, "AAMedianBySample.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=F)


#generate the "Sequences that are present in more than one replicate" dataset
clonaltype.in.replicates = inputdata
clonaltype.in.replicates = clonaltype.in.replicates[clonaltype.in.replicates$Functionality %in% c("productive (see comment)","productive"),]
clonaltype.in.replicates = clonaltype.in.replicates[!(is.na(clonaltype.in.replicates$ID) | is.na(clonaltype.in.replicates$Top.V.Gene) | is.na(clonaltype.in.replicates$Top.J.Gene)),]
clonaltype = unlist(strsplit(clonaltype, ","))

clonaltype.in.replicates$clonaltype = do.call(paste, c(clonaltype.in.replicates[c(clonaltype, "Replicate")], sep = ":"))

clonaltype.in.replicates = clonaltype.in.replicates[!duplicated(clonaltype.in.replicates$clonaltype),]

clonaltype = clonaltype[-which(clonaltype == "Sample")]

clonaltype.in.replicates$clonaltype = do.call(paste, c(clonaltype.in.replicates[clonaltype], sep = ":"))
clonaltype.in.replicates = clonaltype.in.replicates[,c("clonaltype","Replicate", "ID", "Sequence", "Sample")]


write.table(clonaltype.in.replicates, "clonaltypes_replicates_before_table.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)

clonaltype.counts = data.frame(table(clonaltype.in.replicates$clonaltype))

write.table(clonaltype.counts, "clonaltypes_counts.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)

names(clonaltype.counts) = c("clonaltype", "coincidence")

clonaltype.counts = clonaltype.counts[clonaltype.counts$coincidence > 1,]

clonaltype.in.replicates = clonaltype.in.replicates[clonaltype.in.replicates$clonaltype %in% clonaltype.counts$clonaltype,]
clonaltype.in.replicates = merge(clonaltype.in.replicates, clonaltype.counts, by="clonaltype")
clonaltype.in.replicates = clonaltype.in.replicates[order(-clonaltype.in.replicates$coincidence, clonaltype.in.replicates$clonaltype, clonaltype.in.replicates$Replicate),c("coincidence","clonaltype", "Sample", "Replicate", "ID", "Sequence")]


write.table(clonaltype.in.replicates, "clonaltypes_replicates.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)

43:2325074a8461

# ---------------------- AA median CDR3 length ----------------------

median.aa.l = data.frame(data.table(PRODF)[, list(median=as.double(median(as.numeric(.SD$CDR3.Length, na.rm=T), na.rm=T))), by=c("Sample")])
write.table(median.aa.l, "AAMedianBySample.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=F)

if(clonaltype != "none"){
	#generate the "Sequences that are present in more than one replicate" dataset
	clonaltype.in.replicates = inputdata
	clonaltype.in.replicates = clonaltype.in.replicates[clonaltype.in.replicates$Functionality %in% c("productive (see comment)","productive"),]
	clonaltype.in.replicates = clonaltype.in.replicates[!(is.na(clonaltype.in.replicates$ID) | is.na(clonaltype.in.replicates$Top.V.Gene) | is.na(clonaltype.in.replicates$Top.J.Gene)),]
	clonaltype = unlist(strsplit(clonaltype, ","))

	clonaltype.in.replicates$clonaltype = do.call(paste, c(clonaltype.in.replicates[paste_columns], sep = ":"))

	clonaltype.in.replicates = clonaltype.in.replicates[!duplicated(clonaltype.in.replicates$clonaltype),]

	clonaltype = clonaltype[-which(clonaltype == "Sample")]

	clonaltype.in.replicates$clonaltype = do.call(paste, c(clonaltype.in.replicates[clonaltype], sep = ":"))
	clonaltype.in.replicates = clonaltype.in.replicates[,c("clonaltype","Replicate", "ID", "Sequence", "Sample")]


	write.table(clonaltype.in.replicates, "clonaltypes_replicates_before_table.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)

	clonaltype.counts = data.frame(table(clonaltype.in.replicates$clonaltype))

	write.table(clonaltype.counts, "clonaltypes_counts.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)

	names(clonaltype.counts) = c("clonaltype", "coincidence")

	clonaltype.counts = clonaltype.counts[clonaltype.counts$coincidence > 1,]

	clonaltype.in.replicates = clonaltype.in.replicates[clonaltype.in.replicates$clonaltype %in% clonaltype.counts$clonaltype,]
	clonaltype.in.replicates = merge(clonaltype.in.replicates, clonaltype.counts, by="clonaltype")
	clonaltype.in.replicates = clonaltype.in.replicates[order(-clonaltype.in.replicates$coincidence, clonaltype.in.replicates$clonaltype, clonaltype.in.replicates$Replicate),c("coincidence","clonaltype", "Sample", "Replicate", "ID", "Sequence")]


	write.table(clonaltype.in.replicates, "clonaltypes_replicates.txt" , sep="\t",quote=F,na="-",row.names=F,col.names=T)
} else {
	cat("No clonaltype", file="clonaltypes_replicates_before_table.txt")
	cat("No clonaltype", file="clonaltypes_counts.txt")
	cat("No clonaltype", file="clonaltypes_replicates.txt")
}