snakemake output directory

Question

[Peter Chung]

I am new to snakemake. I have a simple quesiton and I can not do the fastqc output into a directory

rule run_fastqc:
    input:
        "data/ERR127302_1_subset.fastq.gz"
    output:
        directory("result/")
    shell:
        "fastqc {input} -o {output} "

I try the command:

snakemake -j 1 -n

The output is:

Building DAG of jobs...
Nothing to be done.

is there any advices ? Thanks.

Answer 1

[Peter Chung]

I can successfully output the result to the result directory.

rule run_fastqc:
    input:
        "data/ERR127302_1_subset.fastq.gz"
    output:
        "result/{sample}_fastqc.html",
        "result/{sample}_fastqc.zip"
    params:
        dir="result"
    shell:
        "fastqc {input} -o {params.dir}"

The html file and the zip will be in the result directory

[进]

it is right answer

Answer 2

[dariober]

Nothing to be done suggests the output, i.e. directory results, exists and is newer than the input so no need to rerun the jobs.

[Peter Chung]

I want the fastqc output file, the html file and the zip file output to result directory but it output nothing.

I tried to change the output in the Snakefile like below:

output:
    expand("result/{sample}_fastqc.html",sample=SAMPLES),
    expand("result/{sample}_fastqc.zip",sample=SAMPLES)

but it still outputs to the data directory, do you know why

[cpad0112]

Did you define rule all?

[Peter Chung]

no, I didn't. what should I input in rule all

rule all:
    input:
        expand("result/{sample}_fastqc.html",sample=SAMPLES),
        expand("result/{sample}_fastqc.zip",sample=SAMPLES)

it stills output at the data directory

[cpad0112]

Try this as shell command: fastqc {input} -q -f fastq -o result/