Snakefile

rule fastqc:
    input:
        read1="raw_data/{id}_R1.fastq.gz",
        read2="raw_data/{id}_R2.fastq.gz"
    output:
        "{id}/fastqc/results.html"
    shell:
        """
        mkdir -p {wildcards.id}/fastqc
        fastqc --outdir="{wildcards.id}/fastqc" {input.read1} {input.read2}
        """

rule align_reads:
    input:
        read1="raw_data/{id}_R1.fastq.gz",
        read2="raw_data/{id}_R2.fastq.gz"
    output:
        "{id}/STAR/results.bam"
    shell:
        """
        STAR --genomeDir "/some/directory/to/reference" \
            --outFileNamePrefix "{wildcards.id}/STAR/results" \
            --readFilesIn {input.read1} {input.read2}
        """

rule quantify_transcripts:
    input:
        bam=rules.align_reads.output
    output:
        isoforms="{id}/rsem/sample.isoforms.results"
    shell:
        """
        rsem-calculate-expression --alignments {input.bam} {wildcards.id}/rsem/sample
        """

IDs, = glob_wildcards("raw_data/{id}_R1.fastq.gz")

rule gather_results:
    input:
        expand(rules.quantify_transcripts.output.isoforms, id=IDs)
    output:
        expression="expression.txt"
    script: "scripts/gather.R"

rule all:
    input:
        rules.gather_results.output,
        expand(rules.fastqc.output, id=IDs)