shell:
    """
      multiqc \
          --interactive \
          -f {params.in_dir} \
          -o {params.out_dir}  > {log} 2>&1
    """

shell:
    """
    # Get the stats on the sequences using seqkit
    seqkit stats {input} > temp.txt

     # Sort the sequence statistics
     (sed -n '1p' temp.txt; awk 'NR>1{{print}}' temp.txt | \
       sort -V -k1,1) > {output} \
       && rm temp.txt
    """

shell:
    "cat {input} > {output}"

shell:
    """
    validate_mapping_file.py \
       -m {input} \
       -o {params.out_dir}
    """

shell:
    """
    seqkit stats {input} > {output} 2> {log}
    """

shell:
    """
    multiple_split_libraries_fastq.py --sampleid_indicator '.fastq' \
         -i {params.reads_dir} -o {params.out_dir} \
         -p {input.parameter_file}
    """

shell:
    """
    # Count all the sequences after split_labrary's quality filtering and concatenation
    count_seqs.py -i {input} > {output}
    """

shell:
    """
    # Split the seqs.fna file into per sample fna files 
    # this step is necessary because usearch fails when you have a lot of samples
    split_sequence_file_on_sample_ids.py  -i {input} -o  {params.out_dir}
    """

shell:
    """

    [ -d {params.out_dir}/{wildcards.sample}/ ] || mkdir -p {params.out_dir}/{wildcards.sample}/
    PROJECT_DIR=`pwd`
    cd {params.out_dir}/{wildcards.sample}/ 

      identify_chimeric_seqs.py \
         -m usearch61 \
         -i ${{PROJECT_DIR}}/{input.fasta} \
         -r {input.reference_database} \
         -o ${{PROJECT_DIR}}/{params.out_dir}/{wildcards.sample}/

    [ -d ${{PROJECT_DIR}}/{params.chimera_dir}/ ] || mkdir -p ${{PROJECT_DIR}}/{params.chimera_dir}/
    # Rename the chimera.txt file so that it will be sample specific
    mv ${{PROJECT_DIR}}/{params.out_dir}/{wildcards.sample}/chimeras.txt \
      ${{PROJECT_DIR}}/{params.chimera_dir}/{wildcards.sample}_chimera.txt && \
    rm -rf  ${{PROJECT_DIR}}/{params.out_dir}/{wildcards.sample}/
    """

shell:
    "cat {input} > {output}"

shell:
    """
    # Filter out the chimeric sequences fom seqs.fna
    filter_fasta.py \
       -s {input.chimeras} \
       -f {input.seqs} \
       -o {output} \
       -n
    """

shell:
    """
    set +u
    {params.conda_activate}
    set -u

    # Count all the sequences after chimera filtering
    count_seqs.py -i {input} > {output}
    """

shell:
    """
    #set +u;source activate qiime1;set -u
    if [ {params.method_class} == "denovo" ];then
    pick_de_novo_otus.py \
        -i {input.seqs} \
        -o {params.out_dir} \
        -p {input.parameter_file} \
        -f

   else

    pick_open_reference_otus.py \
        -i {input.seqs} \
        -r {input.reference_database} \
        -o {params.out_dir} \
        -p {input.parameter_file} \
        -m {params.method} \
        --suppress_align_and_tree \
        -f

    fi
    """   

shell:
    """
    filter_taxa_from_otu_table.py \
        -i {input} \
        -o {output} \
        -n {params.taxa2filter}
    """

shell:
    """
    # Filter-out the really rare otus "The recommended procedure is to discard those OTUs with a
    # number of sequences less than 0.005% of the total number of sequences" (Navas-Molina et al, 2013)
    filter_otus_from_otu_table.py \
        -i {input} \
        -o {output} \
        --min_count_fraction {params.min_freq}
    """

shell:
    """
    biom summarize-table \
        -i {input} \
        -o {output}
    """

    shell:
        """
        #  #--tree_fp
        if [ -d {params.out_dir} ]; then rm -rf {params.out_dir}; fi 

        set +u;source activate qiime1;set -u && \
        core_diversity_analyses.py \
		    -i  {input.otu_table} \
		    -o  {params.out_dir} \
		    --mapping_fp {input.mapping_file} \
		    --sampling_depth {params.depth} \
		    --parameter_fp {params.parameter_file} \
		    --categories {params.category} \
            --nonphylogenetic_diversity
        """

shell:
    """
    {params.conda_activate}
    # ----- Annotate your enzymes, KOs and pathways by adding a description column ------#
    # EC
    add_descriptions.py -i {input.ec} -m EC -o {output.ec}
    # Metacyc Pathway
    add_descriptions.py -i {input.pathway} -m METACYC -o {output.pathway}
    # KO
    add_descriptions.py -i {input.ko} -m KO -o {output.ko} 

    # Unizip the metagenome contribution files - these files describe the micribes contribution the function profiles
    #find {params.outdir} -type f -name "*contrib.tsv.gz" -exec gunzip {{}} \;
    """

shell:
    """
        # Create an empty file
        mkdir -p {params.outdir} && touch {output.ko}
    """

shell:
    """
    # filter out the OTUs of what you are looking for from the given OTU table e.g. human pathogens
    # And convert biom formattted OTU table to tsv 
    filter_taxa_from_otu_table.py \
        -i {input.biom_table} \
        -o {output.biom} -p {input.list2search} && \
    biom convert     \
           -i  {output.biom}   \
           -o  {output.tsv}     \
           --to-tsv     \
           --header-key taxonomy     \
           --output-metadata-id "Consensus Lineage" > {log} 2>&1
    """

shell:
    """
    # Get the pathogenic OTU names
    OTUS=($(cut -f1 {input.tsv_table} | sed -e 1,2d))
    parallel -j {threads} "grep -wE '{{}}' {input.otu_map}  >> {output}" ::: ${{OTUS[*]}} \
        > {log} 2>&1
    """

shell:
    """
    filter_fasta.py \
        -f {input.seqs} \
        -o {output} \
        -m  {input.otu_map} \
        > {log} 2>&1
    """