nf-flu wrapper script

nf-flu.sh

#!/bin/bash

set -euo pipefail

error_handler() {
    echo -e "\n\033[1;31mError on line $1\033[0m"
    # Perform any cleanup or logging here
}

handle_interrupt() {
    echo -e "\n\033[1;31mERROR:\033[1m Script interrupted...\033[0m"
    cleanup
    exit 1
}

# Trap ERR signal to handle errors
trap 'error_handler $LINENO' ERR

# Trap SIGINT and SIGTERM to handle interruptions
trap handle_interrupt SIGINT SIGTERM

error() {
  echo -e "$(date -Is)  \033[1;31mERROR: \033[0m\033[1m$1\033[0m"
}

info() {
  echo -e "$(date -Is)  \033[1;32mINFO: \033[0m\033[1m$1\033[0m"
}

# Default values
TOTAL_CPUS=$(nproc)
FREE_MEM_KB=$(awk '/MemAvailable/ {print $2}' /proc/meminfo)
FREE_MEM_GB=$(echo "scale=2; $FREE_MEM_KB / 1024 / 1024" | bc)
NF_FLU_VERSION=$(curl -s https://api.github.com/repos/CFIA-NCFAD/nf-flu/releases/latest | grep '"tag_name"' | sed -E 's/.*"tag_name": *"([^"]+)".*/\1/')
NCPUS=$TOTAL_CPUS
MEM_TO_USE_GB=$FREE_MEM_GB
REF_DB=""
PLATFORM="nanopore"
PROFILE="docker"
SAMPLESHEET=""
OUTDIR="$(date -I)-results"
DEST=""

FASTA_ZST_URL="https://api.figshare.com/v2/file/download/41415330"
CSV_ZST_URL="https://api.figshare.com/v2/file/download/41415333"
FASTA_ZST_FILE="influenza.fna.zst"
CSV_ZST_FILE="influenza.csv.zst"

download_file() {
    local url=$1
    local output=$2

    if [ ! -f "$output" ]; then
        info "Downloading $output from $url..."
        curl --silent -SLk "$url" -o "$output"
        info "Downloaded $output."
    else
        info "$output already exists. Skipping download."
    fi
}

# Usage instructions
usage() {
    echo "Usage: $0 [options] [-- [options to pass to Nextflow/nf-flu]]"
    echo ""
    echo "Options:"
    echo "  -h                   Show this help message and exit"
    echo "  -i <samplesheet>     Path to the input sample sheet (required)"
    echo "  -o <outdir>          Path to the output directory (default: $OUTDIR)"
    echo "  -d <dest>            Rsync destination for final results (default: $DEST)"
    echo "  -r <ref_db>          Path to reference database (optional)"
    echo "  -p <platform>        Sequencing platform (illumina/nanopore) (default: $PLATFORM)"
    echo "  -P <profile>         nf-flu profile (default: $PROFILE)"
    echo "  -v <nf_flu_version>  nf-flu version (by default, use the latest release: $NF_FLU_VERSION)"
    echo "  -t <cpus>            Number of CPUs to use (default: $TOTAL_CPUS)"
    echo "  -m <memory>          Memory to use in GB (default: ${FREE_MEM_GB} GB)"
    echo "  -f <influenza_fasta> Influenza.fna.zst path (default: ${FASTA_ZST_FILE}; will download from FigShare if file does not exist ($FASTA_ZST_FILE))"
    echo "  -c <influenza_csv>   Influenza.csv.zst path (default: ${CSV_ZST_FILE}; will download from FigShare if file does not exist ($CSV_ZST_URL))"
    echo ""
    echo "Example usage: run nf-flu with all available memory and CPUs and rsync to a NFS path:"
    echo "  $0 -i samplesheet.csv -d /nfs/path/to/outputs/"
    echo ""
    echo "Example usage: run nf-flu with custom arguments to specify latest Rerio Clair3 variant calling model for R10 and Dorado with SUP model:"
    echo "  $0 -i samplesheet.csv -d /nfs/path/to/outputs/ -- --clair3_user_variant_model /data/rerio/clair3_models/r1041_e82_400bps_sup_v500"
    exit 1
}

# Parse command-line options
while getopts ":ht:m:r:p:P:i:o:d:f:c:" opt; do
    case ${opt} in
        h) usage; exit 0 ;;
        t) NCPUS=$OPTARG ;;
        m) MEM_TO_USE_GB=$OPTARG ;;
        r) REF_DB=$OPTARG ;;
        p) PLATFORM=$OPTARG ;;
        P) PROFILE=$OPTARG ;;
        i) SAMPLESHEET=$OPTARG ;;
        o) OUTDIR=$OPTARG ;;
        d) DEST=$OPTARG ;;
        v) NF_FLU_VERSION=$OPTARG ;;
        f) FASTA_ZST_FILE=$OPTARG ;;
        c) CSV_ZST_FILE=$OPTARG ;;
        \?) echo "Invalid option: -$OPTARG" >&2; usage; exit 1 ;;
        :) echo "Option -$OPTARG requires an argument." >&2; usage; exit 1 ;;
    esac
done

shift $((OPTIND - 1))

if [[ "${1:-}" == "--" ]]; then
    shift
fi

# Validate required arguments
if [[ -z "$SAMPLESHEET" || -z "$OUTDIR" ]]; then
    echo -e "\033[1;31mERROR:\033[1m Missing required arguments for samplesheet and nf-flu output directory (-i, -o)\033[0m"
    usage
fi

info "Starting nf-flu (v$NF_FLU_VERSION) ${PLATFORM} analysis with ${NCPUS} CPU cores and ${MEM_TO_USE_GB} GB memory..."

info "Download nf-flu FASTA and CSV files if necessary"
download_file "$FASTA_ZST_URL" "$FASTA_ZST_FILE"
download_file "$CSV_ZST_URL" "$CSV_ZST_FILE"

# Pull the nf-flu pipeline
echo "Pulling the nf-flu pipeline..."
NXF_PULL_CMD="nextflow pull CFIA-NCFAD/nf-flu -r master"
echo "Running: $NXF_PULL_CMD"
eval "$NXF_PULL_CMD" || { echo -e "\033[1;31mERROR:\033[1m Failed to pull nf-flu pipeline\033[0m"; exit 1; }

# Construct the Nextflow command
NF_FLU_CMD="nextflow run CFIA-NCFAD/nf-flu \
  -resume \
  -r "$NF_FLU_VERSION" \
  -profile $PROFILE \
  --platform $PLATFORM \
  --max_cpus $NCPUS \
  --max_memory \"${MEM_TO_USE_GB} GB\" \
  --ncbi_influenza_fasta $FASTA_ZST_FILE \
  --ncbi_influenza_metadata $CSV_ZST_FILE \
  --input $SAMPLESHEET \
  --outdir $OUTDIR $@"

[[ -n "$REF_DB" ]] && NF_FLU_CMD+=" --ref_db $REF_DB"

# Run the nf-flu pipeline
echo "Running the nf-flu pipeline..."
echo "Running: $NF_FLU_CMD"
eval "$NF_FLU_CMD" || { echo -e "\033[1;31mERROR:\033[1m nf-flu pipeline execution failed\033[0m"; exit 1; }

# Sync results to the destination
if [[ -z "$DEST" ]]; then
    echo "No valid rsync destination specified. Destination='$DEST'"
else
    echo "Syncing results to the destination directory..."
    RSYNC_CMD="rsync -rt --progress --stats --exclude=\"work/\" --exclude=\".nextflow/\" \"$PWD\" \"$DEST\""
    echo "Running: $RSYNC_CMD"
    eval "$RSYNC_CMD" || { echo -e "\033[1;31mERROR:\033[1m Failed to sync results to destination\033[0m"; exit 1; }

    echo -e "\033[1;32mSUCCESS:\033[1m nf-flu pipeline completed successfully, and results are synced to $DEST\033[0m"
fi