Load libraries and other scripts

##################
# LOAD LIBRARIES #
##################
suppressWarnings({suppressMessages({suppressPackageStartupMessages({
library(tidyverse)
})  })  })

#########
# PATHS #
#########
input_dir <- "../results/02_data_preprocessing_output/"
result_dir <- "../results/03_normalize_data_output/"
if( isFALSE(dir.exists(result_dir)) ) { dir.create(result_dir,recursive = TRUE) }

#################
# DOWNLOAD DATA #
#################
## TRX DATA:
GEO_url <- "https://www.ncbi.nlm.nih.gov/geo/download/?acc=GSE194276&format=file&file=GSE194276%5FRaw%5Fgene%5Fcounts%5Fmatrix%2Ecsv%2Egz"
# download.file(GEO_url, paste0("../data/"," Raw_gene_counts_matrix.csv.gz"), method="auto")
# system("gunzip ../data/*.gz")

## MICROBIOME DATA:
# if(params$run=="Boston_run1"){
#   url <- c(
#     "ftp://ftp.sra.ebi.ac.uk/vol1/fastq/SRR919/001/SRR9198521/SRR9198521.fastq.gz",
#     "ftp://ftp.sra.ebi.ac.uk/vol1/fastq/SRR919/007/SRR9198517/SRR9198517.fastq.gz",
#     "ftp://ftp.sra.ebi.ac.uk/vol1/fastq/SRR919/008/SRR9198518/SRR9198518.fastq.gz"
#     ) }

# purrr::walk(url, ~download.file(.x, file.path("../data/Fastq/", basename(.x)), method="auto"))
# system("gunzip ../data/*.gz")
# 

#############
# LODA DATA #
#############
trx <- read.csv(paste0("../data/","Raw_gene_counts_matrix.csv"))

dataset_names <- c("ASV_Luminal_raw_counts.csv", # Tissue, Boston run 1 (108 samples)
                   "ASV_Tissue_raw_counts.csv")  # Tissue, Boston run 2+1 (93 sample)

datasets <- map(dataset_names, ~read.csv(paste0(input_dir,.x),row.names = 1)) %>% set_names(., dataset_names)

normalize and save microbiome datasets

# normalized files
datasets <- datasets %>%
  map(., ~round( t(log2( t(.x) / colSums(.x) * 1000 + 1 )),4))

write.csv( datasets[["ASV_Luminal_raw_counts.csv"]] , paste0(result_dir,"ASV_tissue_normalized.csv"),row.names = T )
write.csv( datasets[["ASV_Tissue_raw_counts.csv"]] , paste0(result_dir,"ASV_luminal_normalized.csv"),row.names = T )

normalize and save trx datasets

trx <- rowsum(trx[,-c(1:3)],trx$symbol)
trx  <- trx[ rownames(trx) != "" , ]
trx <- trx[ rownames(trx) != "NA" , ]
trx <- trx[ rowSums(trx > 5) >= 3 , ]
trx <- t( log2( t( trx ) / colSums(trx) * 1e6 + 1 ) )

write.csv( trx , paste0(result_dir,"Tissue_RNAseq_normalized.csv"),row.names = T )

normalized_data <- list( datasets[["ASV_Luminal_raw_counts.csv"]], 
                         datasets[["ASV_Tissue_raw_counts.csv"]], 
                         trx) %>% 
  set_names(., c("ASV_Luminal_normalized", "ASV_Tissue_normalized", "Tissue_RNAseq_V3_normalized"))

saveRDS(normalized_data, paste0(result_dir, "datasets_all_samples.RDS"))