def test_alt_index_in_middle(self):
with open("Roche/E3MFGYR02_alt_index_in_middle.sff", "rb") as handle:
sff2 = list(SffIterator(handle))
self.check_same(sff2)
def test_index_at_start(self):
with open("Roche/E3MFGYR02_index_at_start.sff", "rb") as handle:
sff2 = list(SffIterator(handle))
self.check_same(sff2)
def fileiter(handle):
for record in SffIterator(handle):
# print(record.id)
i = record.id
# Ugly code to make test files...
index = ".diy1.00This is a fake index block (DIY = Do It Yourself), which is allowed under the SFF standard.\0"
padding = len(index) % 8
if padding:
padding = 8 - padding
index += chr(0) * padding
assert len(index) % 8 == 0
# Ugly bit of code to make a fake index at start
index = ".diy1.00This is a fake index block (DIY = Do It Yourself), which is allowed under the SFF standard.\0"
padding = len(index) % 8
if padding:
padding = 8 - padding
index += chr(0) * padding
with open("Roche/E3MFGYR02_random_10_reads.sff", "rb") as handle:
records = list(SffIterator(handle))
with open("Roche/E3MFGYR02_alt_index_at_start.sff", "w") as out_handle:
w = SffWriter(out_handle, index=False, xml=None)
# Fake the header...
w._number_of_reads = len(records)
w._index_start = 0
w._index_length = 0
w._key_sequence = records[0].annotations["flow_key"]
w._flow_chars = records[0].annotations["flow_chars"]
w._number_of_flows_per_read = len(w._flow_chars)
w.write_header()
w._index_start = out_handle.tell()
w._index_length = len(index)
out_handle.seek(0)
w.write_header() # this time with index info
w.handle.write(index)
def test_trim(self):
with open(self.filename, "rb") as handle:
sff_trim = list(SffIterator(handle, trim=True))
self.assertEqual(len(self.sff), len(sff_trim))
for old, new in zip(self.sff, sff_trim):
self.assertEqual(old.id, new.id)
if __name__ == "__main__":
runner = unittest.TextTestRunner(verbosity=2)
unittest.main(testRunner=runner)
if False:
# Ugly code to make test files...
index = ".diy1.00This is a fake index block (DIY = Do It Yourself), which is allowed under the SFF standard.\0"
padding = len(index) % 8
if padding:
padding = 8 - padding
index += chr(0) * padding
assert len(index) % 8 == 0
# Ugly bit of code to make a fake index at start
records = list(
SffIterator(open("Roche/E3MFGYR02_random_10_reads.sff", "rb")))
out_handle = open("Roche/E3MFGYR02_alt_index_at_start.sff", "w")
index = ".diy1.00This is a fake index block (DIY = Do It Yourself), which is allowed under the SFF standard.\0"
padding = len(index) % 8
if padding:
padding = 8 - padding
index += chr(0) * padding
w = SffWriter(out_handle, index=False, xml=None)
# Fake the header...
w._number_of_reads = len(records)
w._index_start = 0
w._index_length = 0
w._key_sequence = records[0].annotations["flow_key"]
w._flow_chars = records[0].annotations["flow_chars"]
w._number_of_flows_per_read = len(w._flow_chars)
w.write_header()
elif keep_negatives:
if len(seq) >= min_len:
negs += 1
yield record
else:
short_neg += 1
in_handle = open(in_file, "rb")
try:
manifest = ReadRocheXmlManifest(in_handle)
except ValueError:
manifest = None
in_handle.seek(0)
out_handle = open(out_file, "wb")
writer = SffWriter(out_handle, xml=manifest)
writer.write_file(process(SffIterator(in_handle)))
# End of SFF code
elif seq_format.lower().startswith("fastq"):
in_handle = open(in_file, "rU")
out_handle = open(out_file, "w")
reader = fastqReader(in_handle)
writer = fastqWriter(out_handle)
if forward:
for record in reader:
seq = record.sequence.upper()
result = primer.search(seq)
if result:
# Forward primer, take everything after it
cut = result.end()
record.sequence = seq[cut:]
if len(record.sequence) >= min_len:
except ImportError:
#Prior to Biopython 1.56 this was a private function
from Bio.SeqIO.SffIO import _sff_read_roche_index_xml as ReadRocheXmlManifest
in_handle = open(in_file, "rb") #must be binary mode!
try:
manifest = ReadRocheXmlManifest(in_handle)
except ValueError:
manifest = None
#This makes two passes though the SFF file with isn't so efficient,
#but this makes the code simple.
pos_count = neg_count = 0
if out_positive_file is not None:
out_handle = open(out_positive_file, "wb")
writer = SffWriter(out_handle, xml=manifest)
in_handle.seek(0) #start again after getting manifest
pos_count = writer.write_file(rec for rec in SffIterator(in_handle)
if clean_name(rec.id) in ids)
out_handle.close()
if out_negative_file is not None:
out_handle = open(out_negative_file, "wb")
writer = SffWriter(out_handle, xml=manifest)
in_handle.seek(0) #start again
neg_count = writer.write_file(rec for rec in SffIterator(in_handle)
if clean_name(rec.id) not in ids)
out_handle.close()
#And we're done
in_handle.close()
#At the time of writing, Galaxy doesn't show SFF file read counts,
#so it is useful to put them in stdout and thus shown in job info.
print "%i with and %i without specified IDs" % (pos_count, neg_count)
elif seq_format.lower() == "fasta":
try:
from Bio.SeqIO.SffIO import ReadRocheXmlManifest
except ImportError:
# Prior to Biopython 1.56 this was a private function
from Bio.SeqIO.SffIO import _sff_read_roche_index_xml as ReadRocheXmlManifest
in_handle = open(in_file, "rb") # must be binary mode!
try:
manifest = ReadRocheXmlManifest(in_handle)
except ValueError:
manifest = None
out_handle = open(out_file, "wb")
writer = SffWriter(out_handle, xml=manifest)
in_handle.seek(0) # start again after getting manifest
count = writer.write_file(rename_seqrecords(SffIterator(in_handle), rename))
out_handle.close()
in_handle.close()
else:
# Use Galaxy for FASTA, QUAL or FASTQ
if seq_format.lower() in ["fasta", "csfasta"] or seq_format.lower().startswith("qual"):
from galaxy_utils.sequence.fasta import fastaReader, fastaWriter
reader = fastaReader(open(in_file, "rU"))
writer = fastaWriter(open(out_file, "w"))
marker = ">"
elif seq_format.lower().startswith("fastq"):
from galaxy_utils.sequence.fastq import fastqReader, fastqWriter
reader = fastqReader(open(in_file, "rU"))
writer = fastqWriter(open(out_file, "w"))
marker = "@"
else: