73 lines (60 with data), 2.3 kB

from singlecellmultiomics.libraryDetection.sequencingLibraryListing import SequencingLibraryLister
from glob import glob
import collections
from singlecellmultiomics.utils import get_contig_list_from_fasta

"""
This workflow:
    Starts off from a folder containing fastq files
    - Detects libraries
    - Concatenates all fastq files belonging to the same library
"""

# This code detects which libraries are present in the current folder:
l = SequencingLibraryLister()
LIBRARIES = l.detect(glob('*.fastq.gz'), merge='_')
# Flatten to library:[fastqfile, fastqfile, ...]
fastq_per_lib = collections.defaultdict(list)

fastq_per_lib_R1 = collections.defaultdict(list)
fastq_per_lib_R2 = collections.defaultdict(list)

for lib,lane_dict in LIBRARIES.items():
    for lane,read_dict in lane_dict.items():
        fastq_per_lib_R1[lib] += read_dict['R1']
        fastq_per_lib_R2[lib] += read_dict['R2']
        fastq_per_lib[lib] += read_dict['R1']
        fastq_per_lib[lib] += read_dict['R2']
libraries =  list( fastq_per_lib.keys() )


def get_fastq_file_list(wildcards):
    # Obtain a list of fastq files associated to wildcards.library
    global libraries
    return sorted( fastq_per_lib[wildcards.library] )

def get_fastq_file_list_R1(wildcards):
    # Obtain a list of fastq files associated to wildcards.library
    global libraries
    return sorted( fastq_per_lib_R1[wildcards.library] )

def get_fastq_file_list_R2(wildcards):
    # Obtain a list of fastq files associated to wildcards.library
    global libraries
    return sorted( fastq_per_lib_R2[wildcards.library] )

def get_target_merge_list():
    global libraries
    targets = []
    for lib in libraries:
        targets.append(f'merged/{lib}_merged_R1.fastq.gz' )
        targets.append(f'merged/{lib}_merged_R2.fastq.gz' )
    return targets

rule all:
    input:
        get_target_merge_list()

rule merge_R1:
    input:
        fastqfiles = get_fastq_file_list_R1
    output:
        merged_fastq = "merged/{library}_merged_R1.fastq.gz"
    shell:
        "cat {input.fastqfiles} > {output.merged_fastq}"

rule merge_R2:
    input:
        fastqfiles = get_fastq_file_list_R2
    output:
        merged_fastq = "merged/{library}_merged_R2.fastq.gz"
    shell:
        "cat {input.fastqfiles} > {output.merged_fastq}"