Snakefile

                "abundance.h5"),
            sample=[i for i in pd.unique(samples_table.index.values)])),
        sample_name_list = ','.join(expand(
            "{sample}",
            sample=pd.unique(samples_table.index.values))),
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--input',
                '--names',
                '--txOut',
                '--output',
                )
            )

    log:
        stderr = os.path.join(
            config["log_dir"],
            current_rule + ".stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            current_rule + ".stdout.log")

    threads: 1

    singularity:
        "docker://zavolab/merge_kallisto:0.6"

    shell:
        "(merge_kallisto.R \
        --input {params.tables} \
        --names {params.sample_name_list} \
        --output {params.dir_out} \
        {params.additional_params}) \
        1> {log.stdout} 2> {log.stderr}"


current_rule = 'pca_salmon'
rule pca_salmon:
    input:
        tpm = os.path.join(
            config["output_dir"],
            "summary_salmon",
            "quantmerge",
            "{molecule}_tpm.tsv"),

    output:
        out = directory(os.path.join(
            config["output_dir"],
            "zpca",
            "pca_salmon_{molecule}"))

    params:
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--tpm',
                '--out',
                )
            )

    log:
        stderr = os.path.join(
            config["log_dir"],
            current_rule + "_{molecule}.stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            current_rule + "_{molecule}.stdout.log")

    threads: 1

    singularity:
        "docker://zavolab/zpca:0.8.3-1"

    shell:
        "(zpca-tpm  \
        --tpm {input.tpm} \
        --out {output.out} \
        {params.additional_params}) \
        1> {log.stdout} 2> {log.stderr}"


current_rule = 'pca_kallisto'
rule pca_kallisto:
    input:
        tpm = os.path.join(
            config["output_dir"],
            "summary_kallisto",
            "{molecule}_tpm.tsv")


    output:
        out = directory(os.path.join(
            config["output_dir"],
            "zpca",
            "pca_kallisto_{molecule}"))

    params:
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--tpm',
                '--out',
                )
            )

    log:
        stderr = os.path.join(
            config["log_dir"],
            current_rule + "_{molecule}.stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            current_rule + "_{molecule}.stdout.log")

    threads: 1

    singularity:
        "docker://zavolab/zpca:0.8.3-1"

    shell:
        "(zpca-tpm  \
        --tpm {input.tpm} \
        --out {output.out} \
        {params.additional_params}) \
        1> {log.stdout} 2> {log.stderr}"


current_rule = 'star_rpm'
rule star_rpm:
    '''
        Create stranded bedgraph coverage with STARs RPM normalisation
    '''
    input:
        bam = lambda wildcards:
            expand(
                os.path.join(
                    config["output_dir"],
                    "samples",
                    "{sample}",
                    "map_genome",
                    "{sample}.{seqmode}.Aligned.sortedByCoord.out.bam"),
                sample=wildcards.sample,
                seqmode=get_sample(
                    'seqmode',
                    search_id='index',
                    search_value=wildcards.sample)),
        bai = lambda wildcards:
            expand(
                os.path.join(
                    config["output_dir"],
                    "samples",
                    "{sample}",
                    "map_genome",
                    "{sample}.{seqmode}.Aligned.sortedByCoord.out.bam.bai"),
                sample=wildcards.sample,
                seqmode=get_sample(
                    'seqmode',
                    search_id='index',
                    search_value=wildcards.sample))

    output:
        str1 = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "STAR_coverage",
            "{sample}_Signal.Unique.str1.out.bg")),
        str2 = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "STAR_coverage",
            "{sample}_Signal.UniqueMultiple.str1.out.bg")),
        str3 = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "STAR_coverage",
            "{sample}_Signal.Unique.str2.out.bg")),
        str4 = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "STAR_coverage",
            "{sample}_Signal.UniqueMultiple.str2.out.bg"))

    shadow: "full"

    params:
        out_dir = lambda wildcards, output:
            os.path.dirname(output.str1),
        prefix = lambda wildcards, output:
            os.path.join(
                os.path.dirname(output.str1),
                str(wildcards.sample) + "_"),
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--runMode',
                '--inputBAMfile',
                '--outWigType',
                '--outFileNamePrefix',
                )
            )

    singularity:
        "docker://quay.io/biocontainers/star:2.7.8a--h9ee0642_1"

    log:
        stderr = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + ".stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + ".stdout.log")

    threads: 4

    shell:
        "(mkdir -p {params.out_dir}; \
        chmod -R 777 {params.out_dir}; \
        STAR \
        --runMode inputAlignmentsFromBAM \
        --runThreadN {threads} \
        --inputBAMfile {input.bam} \
        --outWigType bedGraph \
        --outFileNamePrefix {params.prefix}) \
        {params.additional_params} \
        1> {log.stdout} 2> {log.stderr}"


current_rule = 'rename_star_rpm_for_alfa'
rule rename_star_rpm_for_alfa:
    input:
        plus = lambda wildcards:
            expand(
                os.path.join(
                    config["output_dir"],
                    "samples",
                    "{sample}",
                    "STAR_coverage",
                    "{sample}_Signal.{unique}.{plus}.out.bg"),
                sample=wildcards.sample,
                unique=wildcards.unique,
                plus=get_sample(
                    'alfa_plus',
                    search_id='index',
                    search_value=wildcards.sample)),
        minus = lambda wildcards:
            expand(
                os.path.join(
                    config["output_dir"],
                    "samples",
                    "{sample}",
                    "STAR_coverage",
                    "{sample}_Signal.{unique}.{minus}.out.bg"),
                sample=wildcards.sample,
                unique=wildcards.unique,
                minus=get_sample(
                    'alfa_minus',
                    search_id='index',
                    search_value=wildcards.sample))

    output:
        plus = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.{unique}.plus.bg")),
        minus = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.{unique}.minus.bg"))

    log:
        stderr = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + "_{unique}.stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + "_{unique}.stdout.log")

    singularity:
        "docker://ubuntu:focal-20210416"

    shell:
        "(cp {input.plus} {output.plus}; \
         cp {input.minus} {output.minus};) \
         1>{log.stdout} 2>{log.stderr}"


current_rule = 'generate_alfa_index'
rule generate_alfa_index:
    ''' Generate ALFA index files from sorted GTF file '''
    input:
        gtf = lambda wildcards:
            os.path.abspath(get_sample(
                'gtf',
                search_id='organism',
                search_value=wildcards.organism)),
        chr_len = os.path.join(
            config["star_indexes"],
            "{organism}",
            "{index_size}",
            "STAR_index",
            "chrNameLength.txt"),

    output:
        index_stranded = os.path.join(
            config["alfa_indexes"],
            "{organism}",
            "{index_size}",
            "ALFA",
            "sorted_genes.stranded.ALFA_index"),
        index_unstranded = os.path.join(
            config["alfa_indexes"],
            "{organism}",
            "{index_size}",
            "ALFA",
            "sorted_genes.unstranded.ALFA_index")

    params:
        genome_index = "sorted_genes",
        out_dir = lambda wildcards, output:
            os.path.dirname(output.index_stranded),
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '-a',
                '-g',
                '--chr_len',
                '-o',
                )
            )

    threads: 4

    singularity:
        "docker://quay.io/biocontainers/alfa:1.1.1--pyh5e36f6f_0"

    log:
        os.path.join(
            config["log_dir"],
            current_rule + "_{organism}_{index_size}.log")

    shell:
        "(alfa -a {input.gtf} \
        -g {params.genome_index} \
        --chr_len {input.chr_len} \
        -p {threads} \
        -o {params.out_dir} \
        {params.additional_params}) \
        &> {log}"


current_rule = 'alfa_qc'
rule alfa_qc:
    '''
        Run ALFA from stranded bedgraph files
    '''
    input:
        plus = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.{unique}.plus.bg"),
        minus = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.{unique}.minus.bg"),
        gtf = lambda wildcards:
            os.path.join(
                config["alfa_indexes"],
                get_sample(
                    'organism',
                    search_id='index',
                    search_value=wildcards.sample),
                get_sample(
                    'index_size',
                    search_id='index',
                    search_value=wildcards.sample),
                "ALFA",
                "sorted_genes.stranded.ALFA_index")

    output:
        biotypes = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "ALFA_plots.Biotypes.pdf")),
        categories = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "ALFA_plots.Categories.pdf")),
        table = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.ALFA_feature_counts.tsv")

    params:
        out_dir = lambda wildcards, output:
            os.path.dirname(output.biotypes),
        genome_index = lambda wildcards, input:
            os.path.abspath(
                os.path.join(
                    os.path.dirname(input.gtf),
                    "sorted_genes")),
        plus = lambda wildcards, input:
            os.path.basename(input.plus),
        minus = lambda wildcards, input:
            os.path.basename(input.minus),
        name = "{sample}",
        alfa_orientation = lambda wildcards:
            get_sample(
                'alfa_directionality',
                search_id='index',
                search_value=wildcards.sample),
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '-g',
                '--bedgraph',
                '-s',
                )
            )

    singularity:
        "docker://quay.io/biocontainers/alfa:1.1.1--pyh5e36f6f_0"

    log:
        os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + ".{unique}.log")

    shell:
        "(cd {params.out_dir}; \
        alfa \
        -g {params.genome_index} \
        --bedgraph {params.plus} {params.minus} {params.name} \
        -s {params.alfa_orientation} \
        {params.additional_params}) \
        &> {log}"


current_rule = 'prepare_multiqc_config'
rule prepare_multiqc_config:
    '''
        Prepare config for the MultiQC
    '''
    input:
        script = os.path.join(
            workflow.basedir,
            "workflow",
            "scripts",
            "zarp_multiqc_config.py")

    output:
        multiqc_config = os.path.join(
            config["output_dir"],
            "multiqc_config.yaml")

    params:
        logo_path = config['report_logo'],
        multiqc_intro_text = config['report_description'],
        url = config['report_url'],
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--config',
                '--intro-text',
                '--custom-logo',
                '--url',
                )
            )

    log:
        stderr = os.path.join(
            config["log_dir"],
            current_rule + ".stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            current_rule + ".stdout.log")

    shell:
        "(python {input.script} \
        --config {output.multiqc_config} \
        --intro-text '{params.multiqc_intro_text}' \
        --custom-logo {params.logo_path} \
        --url '{params.url}' \
        {params.additional_params}) \
        1> {log.stdout} 2> {log.stderr}"

current_rule = 'multiqc_report'
rule multiqc_report:
    '''
        Create report with MultiQC
    '''
    input:
        fastqc_se = expand(
            os.path.join(
                config['output_dir'],
                "samples",
                "{sample}",
                "fastqc",
                "{mate}"),
            sample=pd.unique(samples_table.index.values),
            mate="fq1"),

        fastqc_pe = expand(
            os.path.join(
                config['output_dir'],
                "samples",
                "{sample}",
                "fastqc",
                "{mate}"),
            sample=[i for i in pd.unique(
                samples_table[samples_table['seqmode'] == 'pe'].index.values)],
            mate="fq2"),

        pseudoalignment = expand(
            os.path.join(
                config['output_dir'],
                "samples",
                "{sample}",
                "quant_kallisto",
                "{sample}.{seqmode}.kallisto.pseudo.sam"),
            zip,
            sample=[i for i in pd.unique(samples_table.index.values)],
            seqmode=[get_sample('seqmode', search_id='index', search_value=i) 
                for i in pd.unique(samples_table.index.values)]),

        TIN_score = expand(
            os.path.join(
                config['output_dir'],
                "samples",
                "{sample}",
                "TIN",
                "TIN_score.tsv"),
            sample=pd.unique(samples_table.index.values)),

        tables = lambda wildcards:
            expand(
                os.path.join(
                    config["output_dir"],
                    "samples",
                    "{sample}",
                    "ALFA",
                    "{unique}",
                    "{sample}.ALFA_feature_counts.tsv"),
                sample=pd.unique(samples_table.index.values),
                unique=["Unique", "UniqueMultiple"]),

        zpca_salmon = expand(os.path.join(
            config["output_dir"],
            "zpca",
            "pca_salmon_{molecule}"),
            molecule=["genes", "transcripts"]),

        zpca_kallisto = expand(os.path.join(
            config["output_dir"],
            "zpca",
            "pca_kallisto_{molecule}"),
            molecule=["genes", "transcripts"]
        ),

        multiqc_config = os.path.join(
            config["output_dir"],
            "multiqc_config.yaml")

    output:
        multiqc_report = directory(
            os.path.join(
                config["output_dir"],
                "multiqc_summary"))

    params:
        results_dir = os.path.join(
            config["output_dir"]),
        log_dir = config["log_dir"],
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '--outdir',
                '--config',
                )
            )

    log:
        stderr = os.path.join(
            config["log_dir"],
            current_rule + ".stderr.log"),
        stdout = os.path.join(
            config["log_dir"],
            current_rule + ".stdout.log")

    singularity:
        "docker://zavolab/multiqc-plugins:1.2.1"

    shell:
        "(multiqc \
        --outdir {output.multiqc_report} \
        --config {input.multiqc_config} \
        {params.additional_params} \
        {params.results_dir} \
        {params.log_dir};) \
        1> {log.stdout} 2> {log.stderr}"

current_rule = 'sort_bed_4_big'
rule sort_bed_4_big:
    '''
        sort bedGraphs in order to work with bedGraphtobigWig
    '''
    input:
        bg = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "ALFA",
            "{unique}",
            "{sample}.{unique}.{strand}.bg")

    output:
        sorted_bg = temp(os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "bigWig",
            "{unique}",
            "{sample}_{unique}_{strand}.sorted.bg"))

    params:
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=(
                '-i',
                )
            )

    singularity:
        "docker://quay.io/biocontainers/bedtools:2.27.1--h9a82719_5"

    log:
        stderr = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + "_{unique}_{strand}.stderr.log")

    shell:
        "(sortBed \
        -i {input.bg} \
        {params.additional_params} \
        > {output.sorted_bg};) 2> {log.stderr}"


current_rule = 'prepare_bigWig'
rule prepare_bigWig:
    '''
        bedGraphtobigWig, for viewing in genome browsers
    '''
    input:
        sorted_bg = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "bigWig",
            "{unique}",
            "{sample}_{unique}_{strand}.sorted.bg"),
        chr_sizes = lambda wildcards:
            os.path.join(
                config['star_indexes'],
                get_sample(
                    'organism',
                    search_id='index',
                    search_value=wildcards.sample),
                get_sample(
                    'index_size',
                    search_id='index',
                    search_value=wildcards.sample),
                "STAR_index",
                "chrNameLength.txt")

    output:
        bigWig = os.path.join(
            config["output_dir"],
            "samples",
            "{sample}",
            "bigWig",
            "{unique}",
            "{sample}_{unique}_{strand}.bw")

    params:
        additional_params = parse_rule_config(
            rule_config,
            current_rule=current_rule,
            immutable=()
            )

    singularity:
        "docker://quay.io/biocontainers/ucsc-bedgraphtobigwig:377--h0b8a92a_2"

    log:
        stderr = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + "_{unique}_{strand}.stderr.log"),

        stdout = os.path.join(
            config["log_dir"],
            "samples",
            "{sample}",
            current_rule + "_{unique}_{strand}.stdout.log")

    shell:
        "(bedGraphToBigWig \
        {params.additional_params} \
        {input.sorted_bg} \
        {input.chr_sizes} \
        {output.bigWig};) \
        1> {log.stdout} 2> {log.stderr}"