refactor SmartSeq2 inputs

  -c,  --chemistry CHEM         Assay configuration, autodetection is not possible for converted files: 'SC3Pv2' (default), 'SC5P-PE', 'SC5P-R1', 'SC5P-R2', 'threeprime', or 'fiveprime'

                                    5′ scRNA-Seq ('SC5P-PE') is available only for 10x Genomics, ICELL8, SmartSeq, and STRT-Seq technologies.

                                    Setting 'SC3Pv1' for 10x version 1 chemistry is recommended.

                                    All other technologies default to 3′ scRNA-Seq parameters. Only 10x Genomics and ICELL8 allow choosing which to use.

                                    All other technologies default to 3′ scRNA-Seq parameters. Only 10x Genomics, ICELL8, and SmartSeq2 allow choosing which to use.

                                    For SmartSeq2 this parameter detemines using full-length sequences or 5' ends with internal reads removed.

  -n,  --force-cells NUM        Force pipeline to use this number of cells, bypassing the cell detection algorithm.

  -j,  --jobmode MODE           Job manager to use. Valid options: 'local' (default), 'sge', 'lsf', or a .template file

        umiadjust=0

    fi

fi

if [[ "$technology" == "smartseq2" ]] || [[ "$technology" == "smartseq3" ]] || [[ "$technology" == "icell8-5-prime" ]]; then

if [[ "$technology" == "smartseq3" ]] || [[ "$technology" == "icell8-5-prime" ]]; then

    if [[ $verbose ]]; then

        echo "  Using $chemistry for $technology"

    fi

    if [[ "$chemistry" == "fiveprime" ]]; then

       chemistry="SC5P-R1"

       chemistry="SC5P-PE"

    fi

    if [[ "$chemistry" != "SC5P-PE" ]] && [[ "$chemistry" != "SC5P-R1" ]] && [[ "$chemistry" != "SC5P-R2" ]]; then

        echo "Error: option --chemistry must be SC5P-PE, SC5P-R1 or SC5P-R2"

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

            echo "Error: option --chemistry must be SC5P-PE, SC5P-R1 or SC5P-R2 for ${technology} (umi-based)"

            exit 1

        fi

    fi

fi

if [[ "$technology" == "smartseq2"  || ( "$technology" == "smartseq3" && $nonUMI == "true" ) ]]; then

    if [[ $verbose ]]; then

        echo "  Using $chemistry for $technology"

    fi

    if [[ "$chemistry" == "fiveprime" ]]; then

       chemistry="SC5P-PE"

    fi

    if [[ "$chemistry" == "SC5P-R1" ]]; then

        echo "  accurately mapping 5\' ends (filtering by tag sequence)"

    else

        echo "  mapping all reads (tag sequence removed)"

    fi

    if [[ "$chemistry" == "SC5P-PE" ]]; then

        echo "  mapping paired ends..."

    else

        echo "  mapping single-ends..."

    fi

    if [[ "$chemistry" != "SC5P-PE" ]] && [[ "$chemistry" != "SC5P-R1" ]] && [[ "$chemistry" != "SC5P-R2" ]]; then

         echo "  using full-length sequences for read counts (default for ${technology})"

    fi

fi

##########

                }' $convR2 > ${crIN}/.temp

            mv ${crIN}/.temp $convR2

#            echo " ... remove internal reads for ${technology} by matching TSO sequence for UMI reads"

#            #filter UMI reads by matching tag sequence ATTGCGCAATG (bases 1-11 of R1) and remove as an adapters 

#            perl ${FILTERSMARTSEQREADUMI} --r1 ${convR1} --r2 ${convR2} --i1 ${convI1} --i2 ${convI2} --tag 

'AAGCAGTGGTATCAACGCAGAGTAC' --out_dir ${crIN}

#            echo "  ... trim tag sequence from R1"

            if [[ ${chemistry} == "SC5P-R1" ]]; then

                echo " ... remove internal reads for ${technology} by matching TSO sequence for UMI reads"

                #filter UMI reads by matching tag sequence ATTGCGCAATG (bases 1-11 of R1) and remove as an adapters 

                perl ${FILTERSMARTSEQREADUMI} --r1 ${convR1} --r2 ${convR2} --i1 ${convI1} --i2 ${convI2} --tag 'AAGCAGTGGTATCAACGCAGAGTACGG' --out_dir ${crIN}

                echo "  ... trim tag sequence from R1"

                # returns R1 with tag sequence removed (left trim) starting with 8pbp UMI and corresponding reads for I1, I2, and R2

#            mv $crIN/parsed_R1.fastq ${convR1}

#            mv $crIN/parsed_R2.fastq ${convR2}

#            mv $crIN/parsed_I1.fastq ${convI1}

#            mv $crIN/parsed_I2.fastq ${convI2}

                mv $crIN/parsed_R1.fastq ${convR1}

                mv $crIN/parsed_R2.fastq ${convR2}

                mv $crIN/parsed_I1.fastq ${convI1}

                mv $crIN/parsed_I2.fastq ${convI2}

            elif [[ ${chemistry} == "SC3Pv2" ]] || [[ ${chemistry} == "SC5P-PE" ]];

                # remove tag sequence adapter (first occurence only)

                sed -E '

                     /^AAGCAGTGGTATCAACGCAGAGTACGG/ {

                     s/^AAGCAGTGGTATCAACGCAGAGTACGG//g

                     n

                     n

                     s/^.{27}//g

                     }' $convFile > ${crIN}/.temp

                # returns R1 with tag sequence removed  

                mv ${crIN}/.temp $convFile

            fi

            echo "  ... concatencate barcodes to R1 from I1 and I2 index files"

            #concatenate barcocdes from dual indexes to R1 as barcode (bases 1-16)

            ##16 bp barcode, GGG for TSO, no UMI

            mv $crIN/Concatenated_File.fastq ${convR1}

            if [[ $nonUMI == "true" ]]; then

                #add mock UMI (count reads instead of UMI) barcodelength=16, umi_default=10

                perl ${ADDMOCKUMI} --fastq ${convR1} --out_dir ${crIN} --head_length ${barcodelength} --umi_length ${umi_default}

                umilength=${umi_default}

                # returns a combined R1 file with barcode and mock UMI

                ##16 bp barcode, 10 bp UMI, GGG for TSO

                mv $crIN/mock_UMI.fastq ${convR1}

            fi

            # skip adding TSO for 3' chemistry

            if [[ ${chemistry} != "SC3Pv2" ]] && [[ ${chemistry} != "SC3Pv3" ]] && [[ ${chemistry} != "auto" ]]; then

                #convert TSO to expected length for 10x 5' (TSS in R1 from base 39)

                echo " handling $convFile ..."

                tsoS="TTTCTTATATGGG"

                tsoQ="IIIIIIIIIIIII"

                #Add 10x TSO characters to the end of the sequence

#            cmd=$(echo 'sed -E "2~4s/(.{'$barcodelength'})(.{'${umilength}'})(.{3})/\1\2'$tsoS'/" '$convFile' > '${crIN}'/.temp')

                cmd=$(echo 'sed -E "2~4s/(.{'$barcodelength'})(.{'${umilength}'})/\1\2'$tsoS'/" '$convFile' > '${crIN}'/.temp')

                if [[ $verbose ]]; then

                    echo umi: $umilength

                    echo $cmd

                fi

            #run command with barcode and umi length, e.g.,: sed -E "2~4s/(.{16})(.{8})(.{3})(.*)/\1\2$tsoS\4/" $convFile > ${crIN}/.temp

                #run command with barcode and umi length, e.g.,: sed -E "2~4s/(.{16})(.{8})(.*)/\1\2$tsoS\4/" $convFile > ${crIN}/.temp

                eval $cmd

                mv ${crIN}/.temp $convFile

                #Add n characters to the end of the quality

#            cmd=$(echo 'sed -E "4~4s/(.{'$barcodelength'})(.{'${umilength}'})(.{3})/\1\2'$tsoQ'/" '$convFile' > '${crIN}'/.temp')

                cmd=$(echo 'sed -E "4~4s/(.{'$barcodelength'})(.{'${umilength}'})/\1\2'$tsoQ'/" '$convFile' > '${crIN}'/.temp')

            #run command with barcode and umi length, e.g.,: sed -E "4~4s/(.{16})(.{8})(.{3})(.*)/\1\2$tsoQ\4/" $convFile > ${crIN}/.temp

                #run command with barcode and umi length, e.g.,: sed -E "4~4s/(.{16})(.{8})(.*)/\1\2$tsoQ\4/" $convFile > ${crIN}/.temp

                eval $cmd

                mv ${crIN}/.temp $convFile

            fi

            echo "  ${convFile} adjusted"

       done

    fi

                  Fluidigm C1, ICELL8 full-length, inDrops-v3, Quartz-Seq, RamDA-Seq,

                  SCI-RNA-Seq, SCI-RNA-Seq3, scifi-seq, Smart-Seq2, Smart-Seq3, STRT-Seq-2i, STRT-Seq-C1

            For the following full-length technologies the --chemistry argument can be used to refine parameters:

                  SmartSeq, SmartSeq2

  -b,  --barcodefile FILE

            Custom barcode list in plain text (with each line containing a barcode). Please provide

            All samples are converted to contain the barcode and UMI in Read1 as used for 'SC3Pv2'.

            'SC3Pv3' is only used for technologies with longer UMI.

            5′ scRNA-Seq ('SC5P-PE') is available only for 10x Genomics, ICELL8, SmartSeq, and

            5′ scRNA-Seq ('SC5P-PE') is available only for 10x Genomics, ICELL8, SmartSeq3, and

            STRT-Seq technologies. All other technologies default to 3′ scRNA-Seq parameters.

            Only 10x Genomics and ICELL8 allow choosing which chemistry parameter to use.

           For full-length single-cell technologies (SmartSeq and SmartSeq2) the chemistry input

           configures the following settings:

              auto or SC3Pv2 (default): 5' tag sequences are removed and all reads are mapped in 3' chemistry

              SC5P-PE: 5' tag sequences are replaced by the 10x TSO sequence and all reads are mapped in 5' chemistry (paired-ends)

              SC5P-R1: reads not containing a 5' tag are filtered out and 5' ends are mapped using read 1 only

  -n,  --force-cells NUM

            Force pipeline to use this number of cells, bypassing the cell detection algorithm.