Extracting nucleotide sequence from ffn files following Roary

roary2ffn.sh

#!/bin/bash

#########################################
# Anuj Sharma                           #
# [email protected]                      #
# 2022-05-18                            #
#########################################

# Reading the input parameters
OPTIND=1
out_dir="."
unset -v ffn_dir
unset -v geneid_csv
while getopts "d:c:o:hv" opt; do
    case "$opt" in
        d)  ffn_dir=$OPTARG;; # Directory containing ffns.
        c)  geneid_csv=$OPTARG;; # Path of presence/absence CSV.
        o)  out_dir=$OPTARG;;
        h)  echo "Usage: $0  -d ffn_directory  -c presence_absence_csv  [-o output file]" >&2 && exit 0;;
        v)  echo "$(basename $0): v0.2" >&2 && exit 0;;
        *)  echo "Extra arguments not used!" >&2
    esac
done

shift $((OPTIND-1))

# Exit if necessary input are not given
[[ -z "$ffn_dir" && -z "$geneid_csv" ]] && echo "Some input arguments missing." >&2 && exit 1
[[ -z "$ffn_dir" ]] && echo "Input ffn directory path missing." >&2 && exit 1
[[ -z "$geneid_csv" ]] && echo "Input geneID CSV path missing." >&2 && exit 1

# Make output directory.
mkdir -p $out_dir

# Read presence/absence CSV.
# Convert , to ~ is a workaround to prevent error from non-delimiter ,
cat $geneid_csv | sed 's/","/"~"/g' | {
    # Ignore header
    read -r

    # Read row elements into array and iterate over it.
    while IFS='~' read -ra gene
    do
        # Extract name for gene/gene group.
        genename=$(echo "${gene[0]}" | tr -d '"')

        # Parse outfut fasta name from gene name. Remove if existing.
        output="$out_dir/$genename.fasta"; rm -rf $output

        # Parse the gene loci for all genomes
        # Remove carriage return if generated in windows
        genelist=( $( echo "${gene[@]:14}" | tr -d '"' | tr -d '\r' ) )

        # Iterate over individual loci
        for i in "${genelist[@]}"
        do
            # Parse ffn file name from loci name
            input="$ffn_dir/$(echo $i | awk -F ".fasta" '{print $1}').fasta_p.ffn"

            # Extract right sequence from ffn based on seqID
            awk -vRS=">" -vORS="" -vFS="\n" -vOFS="\n" -vID="$i" '(index($1, ID) != 0) {print ">"$0}' $input >> $output
        done
    done
}