Example MetaScope Script

Introduction

The following is an example script utilizing the Bowtie 2 sub-workflow. The demultiplex step is omitted.

library(MetaScope)

Creation of indices

nThreads <- 8
loc <- "/restricted/projectnb/pathoscope/reflib/2023_refseq_bowtie"

targets <- tolower(c("bacteria"))
filters <- c('homo sapiens', 'mus musculus')
inVec <- c(targets, filters)

# make directories
mk_all_dir <- function(ind) {
    ind_ <- stringr::str_replace_all(ind, c(" " = "_"))
    if (!dir.exists(file.path(loc, ind_))) dir.create(file.path(loc, ind_))
    }
sapply(inVec, mk_all_dir)

# Download reference and representative genomes
sapply(inVec, 
       function(x) download_refseq(x, reference = TRUE,
                                   representative = TRUE,
                                   compress = TRUE,
                                   patho_out = FALSE,
                                   # Same directory as created above
                                   out_dir = file.path(
                                   loc, stringr::str_replace_all(x, c(" " = "_")))))

# Make indices
for (ind in inVec) {
  ind_ <- stringr::str_replace_all(ind, c(" " = "_"))
  # Create the reference library index files in the destination directory
  if (ind == "fungi") {
      mk_bowtie_index(ref_dir = file.path(loc, ind_), lib_dir = file.path(loc, "all_indices"), 
                  lib_name = ind_,
                  bowtie2_build_options = "--large-index",
                  threads = nThreads, overwrite = TRUE)
  } else {
      mk_bowtie_index(ref_dir = file.path(loc, ind_), lib_dir = file.path(loc, "all_indices"), 
                      lib_name = paste0(ind_, "rep"), threads = nThreads, overwrite = TRUE)
  }
}

Command-line pre-processing (BASH script)

Note, the variable SGE_TASK_ID refers to an index integer set via a parallel processing job submission system for a computing resource.

threads=8

dataDir=/restricted/projectnb/pathoscope/work/aubrey/metascope_benchmarking/miossec_data
dataDir=/restricted/projectnb/pathoscope/work/aubrey/metascope_benchmarking/miossec_data/species_10/abundance_100000/dominance_1

### MAKE WORKING DIR
workingDir=${SGE_TASK_ID}_tmp_miossec
rm -rf $TMPDIR/$workingDir
mkdir $TMPDIR/$workingDir

# Specify input files
input_line=$(sed "${SGE_TASK_ID}q;d" /restricted/projectnb/pathoscope/work/aubrey/metascope_benchmarking/miossec_data/dat_manifest.txt) 
IFS=$'\t'
read -r sampleName readPath1 readPath2 <<< "$input_line"

### TRIM the reads
# CHANGE CROP AND HEADCROP DEPENDING ON MultiQC ANALYSIS
#java -jar ~/pathoscope/code/other/Trimmomatic-0.36/trimmomatic-0.36.jar PE \
#    -phred33 -threads $threads $reads1 $reads2 \
#    $TMPDIR/$workingDir/reads1.paired.fastq.gz \
#    $TMPDIR/$workingDir/reads1.unpaired.fastq.gz \
#    $TMPDIR/$workingDir/reads2.paired.fastq.gz \
#    $TMPDIR/$workingDir/reads2.unpaired.fastq.gz \
#    SLIDINGWINDOW:4:20 LEADING:3 TRAILING:3 MINLEN:36 \
#    CROP:225 HEADCROP:30

### Run MetaScope
indexDir="/restricted/projectnb/pathoscope/reflib/2023_refseq_bowtie/all_indices"
expTag=$sampleName
outDir="/restricted/projectnb/pathoscope/work/aubrey/metascope_benchmarking/refseq2023_output"
tmpDir="$TMPDIR/$workingDir/"

# Choose filter and targets to download
target="bacteriarep,fungi,viruses"
filter="human_T2T,mus_musculus"

# This is where samtools and R are loaded if necessary
module load samtools
module load R

# Change "run_MetaScope.R" to name of R script saved elsewhere
Rscript --vanilla --max-ppsize=500000 run_MetaScope.R \
${readPath1} ${readPath2} ${indexDir} ${expTag} ${outDir} ${tmpDir} ${threads} \
${target} ${filter}

rm -rf $TMPDIR/$workingDir

R script

# Take in arguments from bash script
args <- commandArgs(trailingOnly = TRUE)

readPath1 <- args[1]
readPath2 <- args[2]
indexDir <- args[3]
expTag <- args[4]
outDir <- args[5]
tmpDir <- args[6]
threads <- args[7]
targets <- stringr::str_split(args[8], ",")[[1]]
filters <- stringr::str_split(args[9], ",")[[1]]

# Time this!
now <- Sys.time()

# Load MetaScope
library(MetaScope)

# Align to targets
do_this <- function(x) stringr::str_replace_all(x, c(" " = "_"))
targets_ <- do_this(targets) 
filters_ <- do_this(filters)

# Identify bt2 params
data(bt2_regular_params)
bt2_params <- bt2_regular_params

target_map <- align_target_bowtie(read1 = readPath1,
                                  read2 = readPath2,
                                  lib_dir = indexDir,
                                  libs =  targets_,
                                  align_dir = tmpDir,
                                  align_file = expTag,
                                  overwrite = TRUE,
                                  threads = threads,
                                  bowtie2_options = paste(bt2_params, "-f"),
                                  quiet = FALSE)

# Align to filters
output <- paste(paste0(tmpDir, expTag), "filtered", sep = ".")
final_map <- filter_host_bowtie(reads_bam = target_map,
                                lib_dir = indexDir,
                                libs = filters_,
                                make_bam = FALSE,
                                output = output,
                                threads = threads,
                                overwrite = TRUE,
                                quiet = FALSE,
                                bowtie2_options = bt2_params)

# MetaScope ID

# your_ENTREZ_KEY <- <Your key here>
Sys.setenv(ENTREZ_KEY = your_ENTREZ_KEY)
metascope_id(final_map, input_type = "csv.gz", aligner = "bowtie2",
             NCBI_key = your_ENTREZ_KEY,
             num_species_plot = 15,
             quiet = FALSE,
             out_dir = outDir)

message(capture.output(Sys.time() - now))

Aubrey Odom

October 30, 2024

Introduction

Creation of indices

Command-line pre-processing (BASH script)

R script