add automatic generation of missing dual index (I1 and I2) files from FASTQ headers

        echo " indexes ${index1[@]} and ${index2[@]} found for ${technology}"

    else

        if [[ $setup == "false" ]]; then

            echo "ERROR: note that ${technology} expects dual indexes: I1 and I2 OR R2 and R3"

            exit 1

            echo "WARNING: note that ${technology} expects dual indexes: I1 and I2 OR R2 and R3"

        fi

    fi

fi

#generate missing indexes if required (generating I1 and I2)

if [[ "$technology" == "indrop-v3" ]] || [[ "$technology" == "sciseq2" ]] || [[ "$technology" == "sciseq3" ]] || [[ "$technology" == "scifiseq" ]] || [[ "$technology" == "smartseq2" ]] ||[[ "$technology" == "smartseq3" ]] || [[ "$technology" == "strt-seq-ci" ]] ; then

     echo "dual indexes I1 and I2 required for $technology"

     if [[ ${#index2[@]} -le 1 ]]; then

         echo " automatically generating I1 and I2 index files from file headers"

         index1=("${read1[@]}")

         index2=("${read1[@]}")

         #for ii in $(seq 1 1 ${#read1[@]}); do

         for ii in ${!read1[@]}; do

             #iterate over read1 inputs

             R1_file=${read1[$(( $ii -1 ))]}

             R2_file=$(echo $R1_file | perl -pne 's/(.*)_R1/$1_R2/' )

             I1_file=$(echo $R1_file | perl -pne 's/(.*)_R1/$1_I1/' )

             I2_file=$(echo $R1_file | perl -pne 's/(.*)_R1/$1_I2/' )

             if [[ $verbose ]]; then

                 echo $R1_file

                 echo $R2_file

                 echo $I1_file

                 echo $I2_file

             fi

             # copies index 1 to next line (1st to 2nd) and deletes 3rd line

             cat $R1_file | sed -E "s/ (.):(.):(.):(.*)\+(.*)$/ \1:\2:\3:\4+\5\n\4/g" | sed "3~5d" > $I1_file

             indexlength=$(($(head $I1_file -n 2 | tail -n 1 | wc -c) -1))

             qualscores=$(seq 1 $indexlength | xargs -I {} printf I)

             if [[ $verbose ]]; then

                 echo index of length $indexlength gives quality score $qualscores

             fi

            sed -i "4~4s/^.*$/${qualscores}/g" $I1_file

            # copies index 2 to next line (1st to 2nd) and deletes 3rd line

            cat $R1_file | sed -E "s/ (.):(.):(.):(.*)\+(.*)$/ \1:\2:\3:\4+\5\n\5/g" | sed "3~5d" >  $I2_file

            index2length=$(($(head $I2_file -n 2 | tail -n 1 | wc -c) -1))

            qualscores2=$(seq 1 $index2length | xargs -I {} printf I)

            if [[ $verbose ]]; then

                echo index2 of length $index2length gives quality score $qualscores2

            fi

            sed -i "4~4s/^.*$/${qualscores2})/g" $I2_file

            index1+=("$I1_file")

            index2+=("$I2_file")

        done

        if [[ $verbose ]]; then

            echo index1: $index1

            echo index2: $index2

        fi

    else

        echo " dual indexes found"

    fi

fi