require(readxl)
require(dplyr)
#set working directory
setwd("/Users/gregg/Desktop/NovaSeq4")
#Read in files
MasterKey <- read.csv(file="./combined01_29_2021_16SandITS_MIDkey.csv", header=TRUE, stringsAsFactors = FALSE, na.strings = "")
NS4 <- read.csv(file="./NovaSeq4SampleList.csv", header=TRUE, stringsAsFactors = FALSE, na.strings = "")
#separate and recombine Samples into longer file
NS4_1 <- subset(NS4, select = c(Sample, Well, Researcher, Alternate.Plate.Name, ITS1, Substrate, Project))
NS4_2 <- subset(NS4, select = c(Sample, Well, Researcher, Alternate.Plate.Name, ITS2, Substrate, Project))
NS4_3 <- subset(NS4, select = c(Sample, Well, Researcher, Alternate.Plate.Name, X16S2, Substrate, Project))
NS4_4 <- subset(NS4, select = c(Sample, Well, Researcher, Alternate.Plate.Name, X16S1, Substrate, Project))
#rename columns to allow for remerging
names(NS4_1)[names(NS4_1) == "ITS1"] <- "midplate"
names(NS4_2)[names(NS4_2) == "ITS2"] <- "midplate"
names(NS4_3)[names(NS4_3) == "X16S2"] <- "midplate"
names(NS4_4)[names(NS4_4) == "X16S1"] <- "midplate"
NS4 <- rbind(NS4_1, NS4_2, NS4_3, NS4_4)
NS4 <- subset(NS4, midplate != "")
#rename columns
names(NS4)[names(NS4) == "Well"] <- "wellposition"
names(NS4)[names(NS4) == "Researcher"] <- "client_name"
names(NS4)[names(NS4) == "Project"] <- "project"
names(NS4)[names(NS4) == "Sample"] <- "samplename"
names(NS4)[names(NS4) == "Alternate.Plate.Name"] <- "plate"
names(NS4)[names(NS4) == "Substrate"] <- "substrate"
names(NS4)[names(NS4) == "Alternate.Plate.Name"] <- "plate"
##add unique key element.  Concatenate midplate and Well
MasterKey$Key <- paste(MasterKey$PlateCode, MasterKey$Position, sep = "_", collapse = NULL)
NS4$Key <- paste(NS4$midplate, NS4$wellposition, sep="_", collapse = NULL)
#reduce masterkey to relevant values
MasterKey <- semi_join(MasterKey, NS4, by="Key")
#reorder by same key
MasterKey <- MasterKey[order(MasterKey$Key),]
NS4 <- NS4[order(NS4$Key),]
#fill in forwardmid and reversemid values
NS4$forwardmid <- MasterKey$forward_mid
NS4$reversemid <- MasterKey$reverse_mid
NS4 <- subset(NS4, select = -c(Key))
#rearrange column order
NS4 <- NS4 %>% select(forwardmid, reversemid, samplename, project, wellposition, plate, midplate, substrate, client_name)
#outputs a csv file
write.csv(NS4, "./NovaSeq4_Demux.csv")