shell:
    "ln -vs {input} {output}"

shell:
    "ln -vs {input} {output}"

shell:
    "for fname in {input} ; do echo $(basename -s \".contigs.fa\" \"${{fname}}\") ; done > {output}"

shell:
    "cat {input} > {output[0]}"

shell:
    "(date && cat {input.read1} > {output.or1} && cat {input.read2} > {output.or2} && date) &> >(tee {log})"

shell:
    "(date && clumpify.sh in={input.r1} in2={input.r2} out={output.odup1} out2={output.odup2} dupedist={config[clumpify][dupedist]} dedupe=t optical=t threads={threads} groups={config[clumpify][groups]} -Xmx{config[clumpify][memory]} && date) &> >(tee {log})"

shell:
    "(date && tail -n +2 {input} > {output} && date) &> {log}"

shell:
    "(date && "
    "jgi_summarize_bam_contig_depths --outputDepth {output.depth} --pairedContigs {output.paired} {input} && date) &> {log}"

shell:
    "(date && scripts/Fasta_to_Scaffolds2Bin.sh -i $(dirname {input.max}) -e fa > {output.maxscaf} && "
    "scripts/Fasta_to_Scaffolds2Bin.sh -i $(dirname {input.met}) -e fa > {output.metscaf} && date) &> {log}"

shell:
    "(date && ln -vs {input} {output} && date) &> >(tee {log})"

run:
    tax=pd.read_csv(params.tax, header=0, sep="\t", index_col=0)
    cov=pd.read_csv(input.cov, header=None, sep="\t")
    cov.columns=['transcript_name', 'coverage']

    # keeping only rows that contain 'Eukary' in the 'full_classification' column
    euks=tax[tax['full_classification'].str.contains("Eukary", na=False)].drop(['counts'], axis=1)
    # keeping only rows that have at least 70% 'max_pid'
    filt_euks=euks.query("max_pid >=70")

    # merging taxonomy with coverage
    merged=filt_euks.merge(cov, how="left", on="transcript_name")
    filt_merged=merged[['full_classification','classification', 'coverage']]

    # Grouping same taxonomy and getting sum of coverage
    final=filt_merged.groupby(['full_classification','classification'], as_index=False)['coverage'].sum()

    # writing to file
    final.to_csv(output[0], index=None, sep="\t")

run:
    # Collecting all files in folder 
    directory=os.path.dirname(input[0])
    os.chdir(directory)

    # verify the path using getcwd() 
    cwd = os.getcwd() 

    # print the current directory 
    print("Current working directory is:", cwd) 

    mylist=[f for f in glob.glob("*.txt")]
    mylist

    # making individual dataframes for each file
    dataframes= [ pd.read_csv( f, header=0, sep="\t", usecols=["full_classification", "coverage"]) for f in mylist ] # add arguments as necessary to the read_csv method

    # Merging all files based on common column
    merged=reduce(lambda left,right: pd.merge(left,right,on='full_classification', how='outer'), dataframes)

    # Giving appropriate column names
    names=['full_classification']+mylist
    new_cols=list(map(lambda x: x.replace('_eukaryotes.txt',''),names))
    merged.columns=new_cols

    # checking if any values are "NA"
    merged.isnull().values.any()
    # if "NA" run the following
    merged.fillna('', inplace=True)

    # Removing rows with all zeroes (0 or 0.0)
    merged.set_index('full_classification', inplace=True)  # first to make first column as rownames
    edited=merged.loc[~(merged==0).all(axis=1)]

    # Writing file without zeroes
    edited.to_csv(output[0], sep='\t', index=True, header=True)

run:
    tax=pd.read_csv(input.tax, header=0, sep="\t", index_col=0)
    cov=pd.read_csv(input.cov, header=None, sep="\t")
    cov.columns=['transcript_name', 'coverage']

    # keeping only rows that contain 'Eukary' in the 'full_classification' column
    euks=tax.drop(['counts'], axis=1)

    # merging taxonomy with coverage
    merged=euks.merge(cov, how="left", on="transcript_name")
    filt_merged=merged[['full_classification','classification', 'coverage']]

    # Grouping same taxonomy and getting sum of coverage
    final=filt_merged.groupby(['full_classification','classification'], as_index=False)['coverage'].sum()

    # writing to file
    final.to_csv(output[0], index=None, sep="\t")

run:
    # Collecting all files in folder
    directory=os.path.dirname(input[0])
    os.chdir(directory)

    # verify the path using getcwd()
    cwd = os.getcwd()

    # print the current directory
    print("Current working directory is:", cwd)

    mylist=[f for f in glob.glob("*ALL.txt")]
    mylist

    # making individual dataframes for each file
    dataframes= [ pd.read_csv( f, sep="\t", usecols=['full_classification', 'coverage']) for f in mylist ] # add arguments as necessary to the read_csv method

    # Merging all files based on common column
    merged=reduce(lambda left,right: pd.merge(left,right,on='full_classification', how='outer'), dataframes)

    # Giving appropriate column names
    names=['full_classification']+mylist
    new_cols=list(map(lambda x: x.replace('_eukulele_all.txt',''),names))
    merged.columns=new_cols

    # checking if any values are "NA"
    merged.isnull().values.any()
    # if "NA" run the following
    merged.fillna('', inplace=True)

    # Removing rows with all zeroes (0 or 0.0)
    merged.set_index('full_classification', inplace=True)  # first to make first column as rownames
    edited=merged.loc[~(merged==0).all(axis=1)]

    # Writing file without zeroes

shell:
    "(date && clumpify.sh in={input.r1} in2={input.r2} out={output.odup1} out2={output.odup2} dupedist={config[clumpify][dupedist]} dedupe=t optical=t threads={threads} groups={config[clumpify][groups]} -Xmx{config[clumpify][memory]} && date) &> >(tee {log})"

shell:
    "(date && awk '{{if ($3 ~ /unclassified/ || $3 ~ /Eukaryota/) print $2}}' {input} > {output} && date) &> >(tee {log})"

shell:
    "(date && cat {input.read1} > {output.or1} && cat {input.read2} > {output.or2} && date) &> >(tee {log})"

shell:
    "(date && clumpify.sh in={input.r1} in2={input.r2} out={output.odup1} out2={output.odup2} dupedist={config[clumpify][dupedist]} dedupe=t optical=t threads={threads} groups={config[clumpify][groups]} -Xmx{config[clumpify][memory]} && date) &> >(tee {log})"

shell:
    "(date && megahit -1 {input.sr1} -2 {input.sr2} --kmin-1pass -m 0.9 --k-list 27,37,47,57,67,77,87 --min-contig-len 1000 -t {threads} -o $(dirname {output})/tmp && "
    "cd $(dirname {output}) && "
    "rsync -avP tmp/ . && "
    "ln -sf final.contigs.fa $(basename {output}) && "
    "rm -rf tmp/ && "
    "date) &> >(tee {log})"

shell:
    "(date && TMPDIR={RESULTS_DIR} coverm make -o {output} -t {threads} -r {input.ref} -c {input.read1} {input.read2} && date) &> >(tee {log})"

shell:
    """
    concoct --coverage_file {input.coverage} --composition_file {input.contigs_cut} -t {threads} -b {output}
    """

shell:
    """
    merge_cutup_clustering.py {input.clustering}/clustering_gt1000.csv > {input.clustering}/clustering_merged.csv
    extract_fasta_bins.py {input.contigs} {input.clustering}/clustering_merged.csv --output_path {output}        
    """

    shell:
        "(date && "
        "while read -r line; do ls $(dirname {input})/*.fa | grep -o \"$line\" ; done < {output.sample} > {output.tmpfile} && "
        "sed 's@^@/work/projects/nomis/metaG_JULY_2020/IMP3/@g' {output.tmpfile} | "
        "sed 's@$@/run1/Preprocessing/mg.r1.preprocessed.fq@g' | "
        "awk -F, '{{print $0=$1\",\"$1}}' | awk 'BEGIN{{FS=OFS=\",\"}} {{gsub(\"r1\", \"r2\", $2)}} 1' | "
        "sed $'1 i\\\\\\n# Read pairs:' {output.reads}"     # using forward-slashes to get `\\\n`

rule metabolic:
    input:
        fa=os.path.join(RESULTS_DIR, "bins/bin_collection.done"),
        reads=rules.prep_metabolic.output
    output:
        directory(os.path.join(RESULTS_DIR, "metabolic_output"))
    log:
        os.path.join(RESULTS_DIR, "logs/metabolic.log")
    conda:
        os.path.join(ENV_DIR, "metabolic.yaml")
    params:
        gtdbtk=config["metabolic"]["db"],
        metabolic=config["metabolic"]["directory"]
    threads:
        config["metabolic"]["threads"]
    message:
        "Running metabolic for all MAGs"

shell:
    "(date && "
    "export GTDBTK_DATA_PATH={params.gtdbtk} && "
    "export PERL5LIB && export PERL_LOCAL_LIB_ROOT && export PERL_MB_OPT && export PERL_MM_OPT && "
    """env PERL5LIB="" PERL_LOCAL_LIB_ROOT="" PERL_MM_OPT="" PERL_MB_OPT="" cpanm Array::Split && """
    "perl {params.metabolic}/METABOLIC-C.pl -t {threads} -in-gn $(dirname {input.fa}) -r {input.reads} -o {output} && "
    "date) &> >(tee {log})"

run:
    bin=pd.read_csv(input[0], sep="\t")
    bin_edited=bin[['selected_by_DASTool', 'classification']]       # selecting columns

run:
    kegg=pd.read_csv(input[0], sep="\t", skiprows=1)
    kegg_edited=kegg[['Geneid', 'Chr']]
    kegg_edited.rename(columns = {'Geneid': 'KEGG', 'Chr': 'Contig'}, inplace=True)
    kegg_contigs=(kegg_edited.assign(Contig = kegg_edited['Contig'].str.split(';')).explode('Contig').reset_index(drop=True))
    kegg_contigs=kegg_contigs.reindex(['Contig','KEGG'], axis=1)
    kegg_contigs.to_csv(output[0], sep="\t", index=False)

run:
    kegg=pd.read_csv(input[0], sep="\t", header=0)
    cov=pd.read_csv(input[1], sep="\t", header=None)
    cov.rename(columns={0: 'Contig', 1: 'Coverage'}, inplace=True)

    length=pd.read_csv(input[2], sep="\t", header=None)
    length.rename(columns={0: 'Contig', 1: 'Length'}, inplace=True)

    tmp=pd.merge(kegg, cov, on='Contig')
    all_merged=pd.merge(tmp, length, on='Contig')
    all_merged.to_csv(output[0], sep="\t", index=False)

run:
    scaffold=pd.read_csv(input[0], sep="\t", header=None)
    scaffold.columns=['Contig', 'Bin']
    scaffold.to_csv(output[0], sep="\t", index=False)

run:
    opened=[]
    for ifile in input:
      df=pd.read_csv(ifile, index_col=None, sep="\t", header=0)
      df['Sample']=re.sub("_gtdbtk.txt", "", os.path.basename(ifile))
      df = df.reindex(['Sample','Bin','Taxa'], axis=1)
      opened.append(df)

    frame=pd.concat(opened, axis=0, ignore_index=True)
    frame.to_csv(output[0], sep="\t", index=False)

run:
    opened=[]
    for ifile in input:
      df=pd.read_csv(ifile, index_col=None, sep="\t", header=0)

run:
    opened=[]
    for ifile in input:
      df=pd.read_csv(ifile, index_col=None, sep="\t", header=0)
      opened.append(df)

    frame=pd.concat(opened, axis=0, ignore_index=True)
    frame.to_csv(output[0], sep="\t", index=False)

run:
    opened=[]
    for ifile in input:
      df=pd.read_csv(ifile, index_col=None, sep="\t", header=0)

script:
    "merge_funcs.R"

shell:
    "cat {params} | awk '{{print $1\",\"$2}}' | sed '@^user@d' | sed '[email protected]@@g' | sed '[email protected]_sub.contigs@@g' | awk '!visited[$0]++' > {output}"

shell:
    "(date && summarize-metabolism --input $(dirname {input.bins}) --output {output.sum} --metadata {input.meta} --summary {output.sum}/results/summarize_metabolism.csv --heatmap {output.sum}/results/summarize_metabolism.pdf --aggregate ON --plotting ON && summarize-metabolism --input $(dirname {input.bins}) --output {output.indiv} --metadata {input.meta} --summary {output.indiv}/results/individual_metabolism.csv --heatmap {output.indiv}/results/individual_metabolism.pdf --plotting ON && date) &> >(tee {log})"

shell:
    "script=$(realpath {params.script}) && cd {params.path} && ${{script}}"

shell:
    "(date && export PATH=$PATH:{params.path} && "
    "MagicLamp.py LithoGenie -bin_dir $(dirname {input.bins}) -bin_ext fa -out {output} -t {threads} --norm && date) &> >(tee {log})" 

shell:
    "(date && ln -vs {input.in1} {output.fout1} && "
    "ln -vs {input.in2} {output.fout2} && date) &> >(tee {log})"

shell:
    "(date && cp -v {input.config} {output.tout1} && "
    "cp -v {input.launcher} {output.tout2} && "
    "cp -v {input.runfile} {output.tout3} && "
    "sed -i 's/\"\$sample\"/{wildcards.sample}/g' {output.tout1} && "
    "sed -i 's/\"\$sample\"/{wildcards.sample}/g' {output.tout2} && "
    "sed -i 's/\"\$sample\"/{wildcards.sample}/g' {output.tout3} && date) &> >(tee {log})"

shell:
    "ln -vs {input} {output}"

shell:
    "ln -vs {input} {output}"

shell:
    "for fname in {input.txt} ; do echo echo \"${{fname}}\"\"    \"$(echo {input.FAA}) ; done > {output}"

shell:
    "(date && python {config[mantis][path]}/ run_mantis -t {input.FAA} --output_folder $(dirname {output}) --mantis_config {input.config} --hmmer_threads {params.cores} --cores {threads} --memory {config[mantis][single_mem]} --kegg_matrix && date) &> >(tee {log})"

shell:
    """(date && sed -n '/##FASTA/q;p' {input} | awk '$3=="CDS"' | awk '{{print $9}}' | awk 'gsub(";.*","")' | awk 'gsub("ID=","")' > {output} && date) &> >(tee {log})"""

shell:
    "(date && python3 ./vibrant/VIBRANT/VIBRANT_run.py -t {threads} -i {input} -folder $(dirname $(dirname {output.viout1})) && date) &> >(tee {log})"

shell:
    "(date && export PATH=$PATH:'/scratch/users/sbusi/tools/miniconda3/envs/vcontact2/bin' && "
    "/scratch/users/sbusi/tools/miniconda3/envs/vcontact2/bin/vcontact2 --force-overwrite --raw-proteins {input.v1} --rel-mode 'Diamond' --proteins-fp {input.v2} --db 'ProkaryoticViralRefSeq94-Merged' --pcs-mode MCL --vcs-mode ClusterONE --c1-bin /home/users/sbusi/apps/vcontact2/cluster_one-1.0.jar --output-dir {output.cout5} && date) &> >(tee {log})"

shell:
    "(date && kraken2 --threads {threads} --db {config[kraken2][db]} --confidence 0.75 {input} --output {output.summary} --report {output.report} && date) &> >(tee {log})"

shell:
    "(date && kaiju2table -e -t {config[kaiju][db]}/{params.nodes} -n {config[kaiju][db]}/{params.names} -r {config[kaiju][rank]} -o {output} {input.files} && date) &> >(tee {log})"

shell:
    "(date && checkv end_to_end -d {config[checkv][db]} {input} $(dirname {output}) -t {threads} && date) &> >(tee {log})"

shell:
    "(date && antismash --cpus {threads} --genefinding-tool none --genefinding-gff3 {input.GFF} --fullhmmer --pfam2go --asf --cb-knownclusters --clusterhmmer --cf-create-clusters {input.FA} --output-dir $(dirname {output}) && date) &> >(tee {log})"