SnakeFile

import pandas as pd
import os

configfile: "config.json"
localrules: all,mkdir

df=pd.read_csv(config["meta_file"], sep='\t', header=0, index_col=0)
sample_ids = list(df.index)
df.index = sample_ids

def get_pair_gz(sample_id):
    dir = config["raw_fastq_gz_dir"]
    return tuple(os.path.join(dir, df.loc[str(sample_id), x]) for x in('fastqfiles',))

def get_adapter(sample_id):
    return df.loc[sample_id]["Adapter1"]

rule all:
    input: 
       expand("{dir}/{sample_id}.sam", dir= config["dir_names"]["mapped_dir"], sample_id= sample_ids)  
    output:
       feature_counts = "countfinal.txt"
    params:
       reference = "rno.gff3"
    shell:
       "featureCounts -T 12 -t miRNA -g Name -s 1 -M -a {params.reference} -o {output} {input} "

rule mkdir:
    output: touch(config["file_names"]["mkdir_done"])
    params: dirs = list(config["dir_names"].values())
    shell: "mkdir -p {params.dirs}"

rule trimmed:
    input:
       rules.mkdir.output,     
       read=lambda wildcards:get_pair_gz(wildcards.sample_id)
    output:
       file1 = config["dir_names"]["trimmed_dir"] + "/{sample_id}.trim.R1.fastq"
    params:
       adapter= lambda wildcards:get_adapter(wildcards.sample_id) 
    shell:
       "cutadapt  -a {params.adapter}  -o {output.file1}  {input.read}"

rule bowtie2:
      input:
        p1=rules.trimmed.output.file1
      output:
        mapped_sam_file = config["dir_names"]["mapped_dir"]+"/{sample_id}.sam"
      params:
        threads = config["params"]["bowtie2"]["threads"],
        reference = config["params"]["bowtie2"]["bowtie2_reference"]
      shell: 
        "bowtie2 --very-sensitive-local  -x {params.reference}  -q {input.p1} -S {output.mapped_sam_file}"