desc="number of threads for panphlan to use")
workflow.add_argument("dbfolder",
default=None,
desc="folder containing database")
workflow.add_argument("filesfile",
default=None,
desc="file with filepaths to run on (relative to input)")
workflow.add_argument("ref", default=None, desc="name of reference db")
workflow.add_argument(
"refs",
default=None,
desc="file with list of references (relative to dbfolder)")
args = workflow.parse_args()
in_files = workflow.get_input_files(".fastq.gz")
out_files = workflow.name_output_files(name=in_files,
tag="panphlan_map",
extension="csv.bz2")
if args.filesfile:
with open(args.filesfile) as f:
in_files = [l.strip() for l in f]
if args.dbfolder:
cmd = "panphlan_map.py -c [reference] -i [depend] -o [target] -p [threads] --i_bowtie2_indexes [db]"
else:
cmd = "panphlan_map.py -c [reference] -i [depend] -o [target] -p [threads]"
if args.ref:
refs = [args.ref]
# create a workflow instance, providing the version number and description
# the version number will appear when running this script with the "--version" option
# the description will appear when running this script with the "--help" option
workflow = Workflow(version="0.1", description="A workflow to run KneadData")
# add the custom arguments to the workflow
workflow.add_argument("kneaddata-db", desc="the kneaddata database", default="/work/code/kneaddata/db/")
workflow.add_argument("input-extension", desc="the input file extension", default="fastq")
workflow.add_argument("threads", desc="number of threads for knead_data to use", default=1)
# get the arguments from the command line
args = workflow.parse_args()
# get all input files with the input extension provided on the command line
in_files = workflow.get_input_files(extension=args.input_extension)
# get a list of output files, one for each input file, with the kneaddata tag
out_files = workflow.name_output_files(name=in_files, tag="kneaddata")
# create a task for each set of input and output files to run kneaddata
workflow.add_task_group(
"kneaddata --input [depends[0]] --output [output_folder] --reference-db [kneaddata_db] --threads [threads]",
depends=in_files,
targets=out_files,
output_folder=args.output,
kneaddata_db=args.kneaddata_db,
threads=args.threads)
workflow.go()
db = pickle.load(bz2.BZ2File(args.pkl_database, 'r'))
marker_to_species={}
for marker,info in db['markers'].items():
if info['clade'] in species_list:
marker_to_species[marker]=info['clade']
# read in the sam file and pull out the reads that align with the markers
with open(task.targets[0].name, "w") as file_handle_write:
with open(task.depends[0].name) as file_handle:
for line in file_handle:
if not line.startswith("@"):
data=line.rstrip().split("\t")
reference=data[SAM_REFERENCE_NAME_INDEX]
if reference in marker_to_species.keys():
seq_id = ";".join([data[SAM_READ_NAME_INDEX],marker_to_species[reference]])
seq = data[SAM_SEQ_INDEX]
file_handle_write.write("\n".join([">"+seq_id,seq])+"\n")
# for each of the input files write the fasta file of reads
for infile in workflow.get_input_files(extension=args.input_tag_extension):
outfile = workflow.name_output_files(infile).replace(args.input_tag_extension,"_metaphlan2_marker_aligned_subset.fasta")
workflow.add_task(
find_reads,
depends=infile,
targets=outfile)
workflow.go()