def main(args):
pipe = Pipeline(args)
fastq = {}
# process the mapping file
with open(pipe.args.map_file.name) as f:
reader = csv.DictReader(f, delimiter="\t")
for r in reader:
fq[r["#SampleID"]] = os.path.join(pipe.inputs_dir,
r['ForwardFastqFile'])
pipe.check_file(fastq[r["#SampleID"]], "Sample file")
# run otu picking; its97, its99
otu = pick_otu.pick_otu(pipe.log.info, pipe.args.phred_quality,
pipe.args.max_bad_run, pipe.args.max_n,
pipe.args.phred_offset, pipe.inputs_dir,
pipe.outputs_dir)
otu.run_split(fastq)
otu_biom = ''
# specify otu picking strategy and run core diversity
if pipe.args.otu_strategy == "de_novo":
otu.run_denovo(pipe.args.ref_db)
o = otu.get_output()
otu_biom = o["otu_table.biom"]
elif pipe.args.otu_strategy == "closed":
otu.run_closed(pipe.args.ref_db)
o = otu.get_output()
otu_biom = o["otu_table.biom"]
else:
otu.run_open(pipe.args.ref_db)
o = otu.get_output()
otu_biom = o["otu_table_mc2_w_tax.biom"]
# make sure the biom file is not empty
if not os.path.getsize(otu_biom) > 0:
pipe.log.info("BIOM file is empty.")
exit(0)
pipe.check_file(otu_biom, "OTU table")
summarize_table.callback(
otu_biom, os.path.join(pipe.outputs_dir, "otu_summary_table.txt"),
False, False)
pipe.exec_cmnd("biom convert -i %s -o %s --to-tsv --header-key taxonomy" %
(otu_biom, os.path.join(pipe.outputs_dir, "otu_table.txt")))
pipe.exec_cmnd(
"biom convert -i %s -o %s --to-json --table-type 'OTU table' --process-obs-metadata sc_separated"
% (os.path.join(pipe.outputs_dir, "otu_table.txt"),
os.path.join(pipe.outputs_dir, "otu_table.v1.biom")))
depth = pipe.get_depth(otu_biom, pipe.args.sampling_depth)
# no sample has the number of reads greater than 10,000
if depth < 0:
pipe.log.info(
"Visualization pipeline will not be run, as there are not enough samples."
)
exit(0)
try:
core = diversity.diversity(pipe.log.info, pipe.args.map_file.name,
pipe.inputs_dir, pipe.outputs_dir)
core.run(depth, otu_biom, "", [])
except Exception:
m = "QIIME core diversity pipeline failed with unknown errors."
pipe.log.error(m)
pipe.log_to_db(job_id=pipe.job_id, stack=traceback.format_exc(), msg=m)
exit(0)
# run additional visualization
if pipe.args.job_id is not None:
r_mod_name = 'datavis16s'
pipe.log.info(
'Loading R module: {r_mod_name}.'.format(r_mod_name=r_mod_name))
pipe.load_R_mod(config.PIPELINES_LOC_ON_WRKR + r_mod_name)
vis = importr("datavis16s")
vis.trygraphwrapper(o["otu_table_mc2_w_tax.biom"],
pipe.outputs_dir,
pipe.args.map_file.name,
logfilename=os.path.join(pipe.outputs_dir,
"logfile.txt"),
FUN="allgraphs",
sampdepth=depth)
# remove the join pair directory
pipe.exec_cmnd("rm -rf %s" % os.path.join(pipe.outputs_dir, "join_pair"))
def main(args):
pipe = Pipeline(args)
db_tree = {
"homd": "/mnt/EFS/dbs/homd/HOMD_16S_rRNA_RefSeq_V15.11.tre",
"sv97": "/mnt/EFS/dbs/SILVA_97/97_otus.tre",
"sv99": "/mnt/EFS/dbs/SILVA_99/99_otus.tre",
"gg97": "/mnt/EFS/dbs/Greengenes_97/97_otus.tree",
"gg99": "/mnt/EFS/dbs/Greengenes_99/99_otus.tree"
}
fastq = {}
# process the mapping file
with open(pipe.args.map_file.name) as f:
reader = csv.DictReader(f, delimiter="\t")
for r in reader:
fastq[r["#SampleID"]] = os.path.join(pipe.inputs_dir,
r['ForwardFastqFile'])
pipe.check_file(fastq[r["#SampleID"]], "Sample file")
# run otu picking; gg99, gg97, sv99, or sv97
otu = pick_otu.pick_otu(pipe.log.info, pipe.args.phred_quality,
pipe.args.max_bad_run, pipe.args.max_n,
pipe.args.phred_offset, pipe.inputs_dir,
pipe.outputs_dir)
otu.run_split(fastq)
otu_tree = ''
otu_biom = ''
# specify otu picking strategy and run core diversity
if pipe.args.otu_strategy == "de_novo":
otu.run_denovo(pipe.args.ref_db)
o = otu.get_output()
otu_tree = o["rep_set.tre"]
otu_biom = o["otu_table.biom"]
elif pipe.args.otu_strategy == "closed":
otu.run_closed(pipe.args.ref_db)
o = otu.get_output()
otu_tree = db_tree[pipe.args.ref_db]
otu_biom = o["otu_table.biom"]
else:
otu.run_open(pipe.args.ref_db)
o = otu.get_output()
otu_tree = o["rep_set.tre"]
otu_biom = o["otu_table_mc2_w_tax_no_pynast_failures.biom"]
# make sure the biom file is not empty
if not os.path.getsize(otu_biom) > 0:
pipe.log.info("BIOM file is empty.")
exit(0)
pipe.check_file(otu_biom, "OTU table")
summarize_table.callback(
otu_biom, os.path.join(pipe.outputs_dir, "otu_summary_table.txt"),
False, False)
pipe.exec_cmnd("biom convert -i %s -o %s --to-tsv --header-key taxonomy" %
(otu_biom, os.path.join(pipe.outputs_dir, "otu_table.txt")))
pipe.exec_cmnd(
"biom convert -i %s -o %s --to-json --table-type 'OTU table' --process-obs-metadata sc_separated"
% (os.path.join(pipe.outputs_dir, "otu_table.txt"),
os.path.join(pipe.outputs_dir, "otu_table.v1.biom")))
depth = pipe.get_depth(otu_biom, pipe.args.sampling_depth)
# no sample has the number of reads greater than 10,000
if depth < 0:
pipe.log.info(
"Visualization pipeline will not be run, as there are not enough samples."
)
exit(0)
try:
core = diversity.diversity(pipe.log.info, pipe.args.map_file.name,
pipe.inputs_dir, pipe.outputs_dir)
core.run(depth, otu_biom, otu_tree)
except Exception:
m = "QIIME core diversity pipeline failed with unknown errors."
pipe.log.error(m)
pipe.log_to_db(job_id=pipe.job_id, stack=traceback.format_exc(), msg=m)
exit(0)
# run picrust analysis
if pipe.args.picrust is True:
pt = picrust.picrust(pipe.log.info, pipe.inputs_dir, pipe.outputs_dir)
pipe.check_file(o["seqs.fna"], "Sequence file")
pt.run(o["seqs.fna"])
# run additional visualization
if pipe.args.job_id is not None:
r_mod_name = 'datavis16s'
pipe.log.info(
'Loading R module: {r_mod_name}.'.format(r_mod_name=r_mod_name))
pipe.load_R_mod(config.PIPELINES_LOC_ON_WRKR + r_mod_name)
vis = importr("datavis16s")
vis.trygraphwrapper(otu_biom,
pipe.outputs_dir,
pipe.args.map_file.name,
logfilename=os.path.join(pipe.outputs_dir,
"logfile.txt"),
FUN="allgraphs",
sampdepth=depth)