initial commit

AnnotationCombine.r

+# Assuming unique info for each tax and func! Check this in input data!
+taxFile <- "SampleID-taxa.txt"
+funcFile <- "SampleID-fct.txt"
+# Get read ids as row headers, taxonomic info as column V2
+taxData <- read.table(file=taxFile, sep="\t", row.names=1, quote="\"")
+# Parse taxonomic info and put in columns
+taxData$Kingdom <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*d__", replacement="", taxData$V2))
+taxData$Phylum <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*p__", replacement="", taxData$V2))
+taxData$Class <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*c__", replacement="", taxData$V2))
+taxData$Order <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*o__", replacement="", taxData$V2))
+taxData$Family <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*f__", replacement="", taxData$V2))
+taxData$Genus <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*g__", replacement="", taxData$V2))
+taxData$Species <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*s__", replacement="", taxData$V2))
+taxData$taxInfo <- TRUE # mark read to have taxonomic info
+# Record percentage info
+taxData$KingdomPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*d__[^;]+;", replacement="", taxData$V2))
+taxData$PhylumPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*p__[^;]+;", replacement="", taxData$V2))
+taxData$ClassPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*c__[^;]+;", replacement="", taxData$V2))
+taxData$OrderPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*o__[^;]+;", replacement="", taxData$V2))
+taxData$FamilyPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*f__[^;]+;", replacement="", taxData$V2))
+taxData$GenusPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*g__[^;]+;", replacement="", taxData$V2))
+taxData$SpeciesPercent <- gsub(pattern=";.*", replacement="", gsub(pattern="^.*s__[^;]+;", replacement="", taxData$V2))
+# drop parsed taxonomic input string
+taxData$V2 <- NULL
+# put NA to entries which were missing
+for (i in 1:7) {
+	taxData[grepl('__', taxData[,i]), i] <- NA
+}
+for (i in 9:15) {
+	taxData[grepl('__', taxData[,i]), i] <- NA
+}
+# get functional info
+funcData <- read.table(file=funcFile, sep="\t", quote="\"")
+#funcData <- read.table(file=funcFile, sep="\t", row.names=1, quote="\"")
+#funcData$Level1 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;", replacement="", funcData$V2))
+funcData$Level2 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;[^;]*;", replacement="", funcData$V2))
+#funcData$Level3 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;[^;]*;[^;]*;", replacement="", funcData$V2))
+#funcData$Level4 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;[^;]*;[^;]*;[^;]*;", replacement="", funcData$V2))
+#funcData$Level5 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;[^;]*;[^;]*;[^;]*;[^;]*;", replacement="", funcData$V2))
+#funcData$Level6 <- gsub(pattern=";.*", replacement="", gsub(pattern="^[^;]*;[^;]*;[^;]*;[^;]*;[^;]*;[^;]*;", replacement="", funcData$V2))
+funcData$funcInfo <- TRUE # mark read to have functional info
+# put NA to entries which were missing
+funcData[funcData==" eggNOG"] <- ""
+funcData[funcData=="eggNOG"] <- ""
+funcData[funcData==""] <- NA
+# drop parsed taxonomic input string
+funcData$V2 <- NULL
+
+# if reads with only taxonomic or functional info are of interest:
+#data <- merge(taxData, funcData, by = "row.names", all=TRUE)
+# if only reads are of interest which have both taxonomic info and functional info use:
+data <- merge(taxData, funcData, by = "row.names")
+write.table(data, "datatxt.txt", sep="\t")
+# lets get a table combining the levels of interest
+# Select the taxa and eggnog level of interest (here exemplary: Species and EggNOG Level1)
+myTable <- table(data$Species, data$Level1, useNA="always")
+
+#rownames(myTable)[nrow(myTable)] <- "unclassified"
+#colnames(myTable)[ncol(myTable)] <- "unclassified"
+
+write.table(myTable, "myTable_sampleID_taxa-vs-fct.txt", sep = "\t")