run:
    import re
    from pandas import read_csv
    # read in data
    counts = read_csv(input.counts, sep='\t', header=0, index_col=0) # sample x asv
    tax    = read_csv(input.tax,    sep='\t', header=0, index_col=0) # asv x taxonomy
    # create output files
    for sid in counts.index:
        sid_re = re.compile(".*\.krona\.%s\.txt$" % sid) # pattern should match the output file pattern
        with open(list(filter(sid_re.match, output))[0], "w") as ofile: # matching output file
            for asv in counts.columns:
                if counts.loc[sid, asv] > 0:
                    assert asv in tax.index
                    c = counts.loc[sid, asv] # count
                    t = "\t".join(tax.loc[asv].fillna(value="NA")) # taxonomy
                    t = re.sub("^NA(\tNA)+$", "Unknown", t) # replace completely unknown taxonomy
                    t = re.sub("(\tNA)+$", "", t) # remove trailing NAs
                    ofile.write("%d\t%s\n" % (c, t))

shell:
    "ktImportText {input} -o {output} && sed -i 's/ASV\.krona\.//g' {output}"

shell:
    "fastqc -q -f fastq -t {threads} -o $(dirname {output.zip}) {input} &> {log}"

shell:
    "multiqc --interactive -p -f -m fastqc -o $(dirname {output.html}) $(dirname {input[0]}) &> {log}"

shell:
    "multiqc --interactive -p -f -m cutadapt -o $(dirname {output.html}) $(dirname {input[0]}) &> {log}"