#  -*- html -*-

# Recipe example for unpaired amplicon data (pairs are supported). Steps are
# run in the order they are given. Recipes are run with the run_recipe script
# (type its name to see options) and steps available for recipes are listed 
# with help_recipe.

<recipe>

    title = Pig Staph 16S

    author = Carmen 
    # email = user@company.com
    site = KU

    # -------------------------------------------------------------------------
    #                         SFF -> FASTQ CONVERSION
    # -------------------------------------------------------------------------

    <sequence-conversion>
        title = Conversion to fastq
    </sequence-conversion>
    
    # -------------------------------------------------------------------------
    #                             DE-MULTIPLEXING
    # -------------------------------------------------------------------------

    # Matches one or more primer motifs against the sequences and splits them
    # into separate files, one for each barcode. The barcode is looked for just
    # upstream of the motif.

    <sequence-demultiplex>
        title = Pattern de-multiplexing
        bar-file = XXXXXXXXXXX
        bar-spacing = 2
        bar-quality = 99%
    	primer-file = Pig.primers
    </sequence-demultiplex>

    # -------------------------------------------------------------------------
    #                             BASIC CLEANING
    # -------------------------------------------------------------------------

    <sequence-cleaning>

        title = Sequence cleaning
        # quality-type = Sanger
        # keep-outputs = no

        # ---------------------------------------------------------------------
        #                        PRIMER REMOVALS
        # ---------------------------------------------------------------------

        # The primer demultiplex step above (see the Test.primers file) removed
        # the forward primer, but the end of sequence may include the reverse
        # primer. Here we clip and trim that,

        <sequence-clip-pattern-end>
            title = Reverse 3 end clip
            pattern-string = CCGTCAATTCMTTTR[1,1,1] AGT
            pattern-orient = reverse
            include-match = no
            search-distance = 100
        </sequence-clip-pattern-end>

        <sequence-trim-end>
            title = Reverse 3 end trim
            sequence = ACTYAAAKGAATTGACGG
            search-distance = 100
            minimum-length = 1
            minimum-strict = 80%
        </sequence-trim-end>
	
        # ---------------------------------------------------------------------
        #                          QUALITY TRIM
        # ---------------------------------------------------------------------

        # This removes low quality from beginnings and ends. A window of set
        # length is used, within which a given number of bases must match with
        # a given minimum quality percentage,

        <sequence-trim-quality-start>
            title = Start quality trim
            window-length = 15
            window-match = 14
            minimum-quality = 97
        </sequence-trim-quality-start>

        <sequence-trim-quality-end>
            title = End quality trim
            window-length = 35
            window-match = 34
            minimum-quality = 96
        </sequence-trim-quality-end>

        # ---------------------------------------------------------------------
        #                           LENGTH FILTER
        # ---------------------------------------------------------------------

        # Unlike trimming, filtering discards the whole sequence if sequences
        # are too short after having passed through the steps above,

        <sequence-filter>
            title = Length filter
            minimum-length = 180
        </sequence-filter>

        # ---------------------------------------------------------------------
        #                           QUALITY FILTER
        # ---------------------------------------------------------------------

        # This discards the whole sequence if less than a set percentage has a
        # quality below a given percentage,

        <sequence-filter-quality>
            title = Quality filter
            minimum-quality = 95
            minimum-strict = 90
        </sequence-filter-quality>

    </sequence-cleaning>

    # -------------------------------------------------------------------------
    #                              DE-REPLICATION
    # -------------------------------------------------------------------------

    <sequence-dereplication>
        title = Sequence uniqification
        # keep-outputs = no
    </sequence-dereplication>
    
    # -------------------------------------------------------------------------
    #                            CHIMERA FILTER
    # -------------------------------------------------------------------------

    <sequence-chimera-filter>
        dataset-name = RDP_SSU_domain_3-5-Default
        title = Chimera filtering
        word-length = 8
        step-length = 4
        minimum-score = 30
        # debug-output = yes
        # keep-outputs = no
    </sequence-chimera-filter>

    # -------------------------------------------------------------------------
    #                            RDP PROFILE
    # -------------------------------------------------------------------------

    <sequence-similarities-simrank>
        title = RDP similarities
        output-name = org_seqs
        dataset-names = RDP_SSU_domain_3-5-Default
        match-word-length = 8
        match-step-length = 2
        minimum-base-quality = 95%
        match-minimum = 70%
        match-top-range = 1%
        match-forward = yes
        match-reverse = no
        # keep-outputs = no
    </sequence-similarities-simrank>

    <organism-taxonomy-profiler>
        title = RDP taxonomy profiling
        output-name = org_profile
        taxonomy-names = RDP_SSU
        minimum-oligo-count = 1
        maximum-ambiguity-depth = 1
        map-method = branch_tree
        match-minimum = 70%
        match-use-range = 0%
        score-unclassified-sister = no
        id-file = XXXXXXXXXX
        # sequence-outputs = phylum
        # keep-outputs = no
    </organism-taxonomy-profiler>

    <organism-profile-format>
        title = RDP taxonomic profiles
        input-step = organism-taxonomy-profiler
        output-name = org_profile_rdp
        taxonomy-minimum-score = 5
        zip-outputs = yes
    </organism-profile-format>

</recipe>