tmpdir = args.out.split('.')[0] + '_' + str(os.getpid())
if not os.path.exists(tmpdir):
os.makedirs(tmpdir)
#split fastq file
amptklib.split_fastq(SeqIn, orig_total, tmpdir, cpus * 2)
#now get file list from tmp folder
file_list = []
for file in os.listdir(tmpdir):
if file.endswith(".fq"):
file = os.path.join(tmpdir, file)
file_list.append(file)
#finally process reads over number of cpus
amptklib.runMultiProgress(processRead, file_list, cpus)
print "-------------------------------------------------------"
#Now concatenate all of the demuxed files together
amptklib.log.info("Concatenating Demuxed Files")
tmpDemux = args.out + '.tmp.demux.fq'
with open(tmpDemux, 'wb') as outfile:
for filename in glob.glob(os.path.join(tmpdir, '*.demux.fq')):
if filename == tmpDemux:
continue
with open(filename, 'rU') as readfile:
shutil.copyfileobj(readfile, outfile)
#parse the stats
finalstats = [0, 0, 0, 0, 0, 0, 0]
for file in os.listdir(tmpdir):
stripPrimer(args.fasta)
else:
amptklib.log.info("Using %i cpus to process data" % cpus)
#now split it into chunks (as many cpus as are queried)
amptklib.split_fasta(args.fasta, folder, cpus * 2)
#get list of files
file_list = []
for file in os.listdir(folder):
if file.endswith(".fasta"):
file = os.path.join(folder, file)
file_list.append(file)
#finally process reads over number of cpus
amptklib.runMultiProgress(stripPrimer, file_list, cpus)
#now concatenate outputs together
OutName = args.out + '.extracted.fa'
ErrorName = args.out + '.errors.fa'
with open(OutName, 'w') as outfile:
with open(ErrorName, 'w') as outfile2:
for filename in os.listdir(os.path.join(folder)):
if filename.endswith('.extracted.fa'):
if filename == OutName:
continue
with open(os.path.join(folder, filename), 'rU') as readfile:
shutil.copyfileobj(readfile, outfile)
if filename.endswith('.errors.fa'):
if filename == ErrorName:
continue
sys.exit(1)
#reverse comp primers for search
RevPrimer = revcomp_lib.RevComp(args.rev_primer)
RevForPrimer = revcomp_lib.RevComp(args.fwd_primer)
#setup tmpdir
tmpdir = args.out.split('.')[0] + '_' + str(os.getpid())
if not os.path.exists(tmpdir):
os.makedirs(tmpdir)
#first run forwards
print("Splitting forward reads into buckets")
GoodFor = args.out + '.fwd.stripped.fq'
BadFor = args.out + '.fwd.no_primer.fq'
filelist = splitter(args.input, tmpdir)
print("Stripping primers from forward reads")
amptklib.runMultiProgress(primerStrip, filelist, cpus)
combiner(tmpdir, ".good", GoodFor)
combiner(tmpdir, ".bad", BadFor)
shutil.rmtree(tmpdir)
#now run reverse
print("Splitting reverse reads into buckets")
tmpdir = args.out.split('.')[0] + '_' + str(os.getpid() + 1)
os.makedirs(tmpdir)
GoodRev = args.out + '.rev.stripped.fq'
BadRev = args.out + '.rev.no_primer.fq'
revfilelist = splitter(args.reverse, tmpdir)
print("Stripping primers from reverse reads")
amptklib.runMultiProgress(revprimerStrip, revfilelist, cpus)
combiner(tmpdir, ".good", GoodRev)
combiner(tmpdir, ".bad", BadRev)