Snakemake：仅扩展通配符的子集

Question

I can't find the solution for this probably easy problem:

I have this snakefile, which first produces the following files:

• data/sample1_P1.txt

• data/sample1_P2.txt

• data/sample2_P1.txt

• data/sample2_P2.txt

In the next step, it just concatenates the files to one file concatenated/concatenated.txt .

This is the minimal, reproducible example:

pairs = {"P1" : "P1", "P2" : "P2"}

samples = {
    "sample1": "sample1",
    "sample2": "sample2"
}

rule all:
    input: "concatenated/concatenated.txt"

rule get_txt_files:
    output:
        "data/{sample}_{pair}.txt"
    shell:
        """
        echo 1 > {output}
        """

rule concatenate:
  input:
    expand("data/{sample}_{pair}.txt", sample=samples, \
        pair=pairs)
  output:
    "concatenated/concatenated.txt"
  shell:
    "cat {input} > {output};"

My question is simple: How can I modify the rule concatenate , so that it concatenates the files with the same sample name?

Desired output would be:

• concatenated/sample1.txt

• concatenated/sample2.txt

Any help would be appreciated.

EDIT

I have a very similar follow-up question, so I don't think it's necessary to open a new question again:

What if my expected output would be as follows:

• data/sample1/sample1_P1

• data/sample1/sample1_P2

• data/sample2/sample2_P1

• data/sample2/sample2_P2

To be clear: I only want to create a new direcotry and move the files into that bespoke direcoty.

It seemed intuitive to do it like this:

pairs = {"P1" : "P1", "P2" : "P2"}

samples = {
    "sample1": "sample1",
    "sample2": "sample2"
}

rule all:
    input: expand("data/{sample}/{sample}_{pair}.txt", sample=samples, pair = pairs)

rule get_txt_files:
    output:
        "data/{sample}_{pair}.txt"
    shell:
        """
        echo 1 > {output}
        """

rule reorganise:
  input:
    expand("data/{{sample}}_{pair}.txt", \
        pair=pairs)
  output:
    "data/{sample}/{sample}_{pair}.txt"
  shell:
    "mv {input} data/{wildcards.sample}/.;"

Can you spot the problem?

Thanks a lot in advance

Answer 1

rule concatenate:
  input:
    expand("data/{{sample}}_{pair}.txt", pair=pairs)
  output:
    "concatenated/{sample}.txt"
  shell:
    "cat {input} > {output};"

Answer to q in comment:

from snakemake.io import expand # automatically imported in Snakemake 

expand("data/{{sample}}_{pair}.txt", pair="A B C".split())
# ['data/{sample}_A.txt', 'data/{sample}_B.txt', 'data/{sample}_C.txt']