def stats(fastq_file):
"""Generate basic stats from FASTQ file
"""
# Loop over all reads in the FASTQ
n_reads = 0
read_lengths = {}
index_sequences = {}
for read in FASTQFile.FastqIterator(fastq_file):
# Count of reads
n_reads += 1
# Read length distribution
read_len = len(read.sequence)
if read_len in read_lengths:
read_lengths[read_len] += 1
else:
read_lengths[read_len] = 1
# Tag name distribution
index_seq = read.seqid.index_sequence
if index_seq is not None:
if index_seq in index_sequences:
index_sequences[index_seq] += 1
else:
index_sequences[index_seq] = 1
# Finished
print "Total reads: %d" % n_reads
print "Read lengths"
for len_ in read_lengths:
print "\t%d: %d" % (len_, read_lengths[len_])
print "Index sequences"
for seq in index_sequences:
print "\t%s: %d" % (seq, index_sequences[seq])
def edit_instrument_name(fastq_file,new_instrument_name):
"""Edit the instrument name for all records in FASTQ file
Loop over all records in a supplied FASTQ file, update the sequence identifier
(i.e. first line in the each record) by changing the instrument name, and write
the updated records to stdout.
"""
# Loop over all reads in the FASTQ
# Update the instrument name in the sequence identifier and echo to stdout
for read in FASTQFile.FastqIterator(fastq_file):
if new_instrument_name:
# Modify the instrument name
read.seqid.instrument_name = new_instrument_name
# Echo updated read to stdout
print read
fastqs = sample.fastq_subset(read_number=1) + \
sample.fastq_subset(read_number=2)
for fastq in fastqs:
print "\t\t%s" % fastq
# Report the names of the samples in each project
if options.report:
for project in illumina_data.projects:
print "%s" % describe_project(project)
# Report statistics for fastq files
if options.stats:
# Print number of reads for each file, and file size
for sample in project.samples:
for fastq in sample.fastq:
fq = os.path.join(sample.dirn, fastq)
nreads = FASTQFile.nreads(fq)
fsize = os.path.getsize(fq)
print "%s\t%s\t%d" % (
fastq, bcf_utils.format_file_size(fsize), nreads)
print ""
# Summary: short report suitable for logging file
if options.summary:
print "%s" % summarise_projects(illumina_data)
# Print number of undetermined reads
if options.stats and illumina_data.undetermined is not None:
print "Undetermined indices"
for lane in illumina_data.undetermined.samples:
for fastq in lane.fastq:
fq = os.path.join(lane.dirn, fastq)
"N_SUBSET reads. (Quicker than using all reads but may not be accurate "
"if subset is not representative of the file as a whole.)")
# Process the command line
options, arguments = p.parse_args()
if len(arguments) != 1:
p.error("input FASTQ file required")
else:
fastq_file = arguments[0]
if not os.path.exists(fastq_file):
p.error("Input file '%s' not found" % fastq_file)
# Get broad format type
print "Sniffing %s" % fastq_file
print "\nData from first read:"
for read in FASTQFile.FastqIterator(fastq_file):
fastq_format = read.seqid.format
if fastq_format is None and read.is_colorspace:
fastq_format = 'colorspace'
print "\tHeader format:\t%s" % str(fastq_format)
print "\tSeq length:\t%d" % read.seqlen
break
# Determine the quality score range (and count reads)
try:
n_subset = int(options.n_subset)
except TypeError:
n_subset = None
n_reads = 0
min_max_qual = (None, None)
for read in FASTQFile.FastqIterator(fastq_file):
#######################################################################
# Main program
#######################################################################
if __name__ == "__main__":
# Create command line parser
p = optparse.OptionParser(usage="%prog OPTIONS R1.fastq R2.fastq",
version="%prog "+__version__,
description="Check that read headers for R1 and R2 fastq files "
"are in agreement, and that the files form an R1/2 pair.")
# Parse command line
options,args = p.parse_args()
# Get data directory name
if len(args) != 2:
p.error("expected two arguments (R1 and R2 fastq files to compare)")
fastq_file_r1 = args[0]
fastq_file_r2 = args[1]
# Process the data
if FASTQFile.fastqs_are_pair(fastq_file_r1,fastq_file_r2):
sys.exit(0)
else:
logging.error("Not R1/R2 pair")
sys.exit(1)
fastqs = sample.fastq_subset(read_number=1) + \
sample.fastq_subset(read_number=2)
for fastq in fastqs:
print "\t\t%s" % fastq
# Report the names of the samples in each project
if options.report:
for project in illumina_data.projects:
print "%s" % describe_project(project)
# Report statistics for fastq files
if options.stats:
# Print number of reads for each file, and file size
for sample in project.samples:
for fastq in sample.fastq:
fq = os.path.join(sample.dirn,fastq)
nreads = FASTQFile.nreads(fq)
fsize = os.path.getsize(fq)
print "%s\t%s\t%d" % (fastq,
bcf_utils.format_file_size(fsize),
nreads)
print ""
# Summary: short report suitable for logging file
if options.summary:
print "%s" % summarise_projects(illumina_data)
# Print number of undetermined reads
if options.stats and illumina_data.undetermined is not None:
print "Undetermined indices"
for lane in illumina_data.undetermined.samples:
for fastq in lane.fastq:
def demultiplex_fastq(fastq_file, barcodes, nmismatches):
"""Perform demultiplexing of a FASTQ file
Demultiplex reads in a FASTQ file given information about a set of
barcode/index sequences.
Produces a file for each barcode, plus another for 'unbinned'
reads.
Arguments:
fastq_file: FASTQ file to be demultiplexed (can be gzipped)
barcodes: list of barcode sequences to use for demultiplexing
nmismatches: maxiumum number of mismatched bases allowed when
testing whether barcode sequences match
Returns:
No return value
"""
# Start
print "Processing %s" % fastq_file
info = IlluminaData.IlluminaFastq(fastq_file)
# Set up output files
output_files = {}
# Weed out barcodes that aren't associated with this lane
local_barcodes = []
for barcode in barcodes:
if barcode['lane'] != info.lane_number:
continue
local_barcodes.append(barcode)
output_file_name = "%s_%s_L%03d_R%d_%03d.fastq" % (
barcode['name'], barcode['index'], info.lane_number,
info.read_number, info.set_number)
print "\t%s\t%s" % (barcode['index'], output_file_name)
if os.path.exists(output_file_name):
print "\t%s: already exists,exiting" % output_file_name
sys.exit(1)
output_files[barcode['index']] = open(output_file_name, 'w')
# Check if there's anything to do
if len(local_barcodes) == 0:
return
# Also make a file for unbinned reads
unbinned_file_name = "unbinned_L%03d_R%d_%03d.fastq" % (
info.lane_number, info.read_number, info.set_number)
if os.path.exists(unbinned_file_name):
print "\t%s: already exists,exiting" % unbinned_file_name
sys.exit(1)
output_files['unbinned'] = open(unbinned_file_name, 'w')
# Process reads
nreads = 0
for read in FASTQFile.FastqIterator(fastq_file):
nreads += 1
matched_read = False
this_barcode = read.seqid.index_sequence
for barcode in local_barcodes:
if barcode['matcher'].match(this_barcode, nmismatches):
##print "Matched %s against %s" % (this_barcode,barcodes[barcode]['name'])
output_files[barcode['index']].write(str(read) + '\n')
matched_read = True
break
# Put in unbinned if no match
if not matched_read:
output_files['unbinned'].write(str(read) + '\n')
##if nreads > 100: break
# Close files
for barcode in local_barcodes:
output_files[barcode['index']].close()
print "\tMatched %d reads for %s" % (nreads, os.path.basename(fastq_file))
os.path.normpath(os.path.join(os.path.dirname(sys.argv[0]), '..',
'share')))
sys.path.append(SHARE_DIR)
import FASTQFile
#######################################################################
# Main program
#######################################################################
if __name__ == "__main__":
# Create command line parser
p = optparse.OptionParser(
usage="%prog OPTIONS R1.fastq R2.fastq",
version="%prog " + __version__,
description="Check that read headers for R1 and R2 fastq files "
"are in agreement, and that the files form an R1/2 pair.")
# Parse command line
options, args = p.parse_args()
# Get data directory name
if len(args) != 2:
p.error("expected two arguments (R1 and R2 fastq files to compare)")
fastq_file_r1 = args[0]
fastq_file_r2 = args[1]
# Process the data
if FASTQFile.fastqs_are_pair(fastq_file_r1, fastq_file_r2):
sys.exit(0)
else:
logging.error("Not R1/R2 pair")
sys.exit(1)
