def test_barcode_size_diff(self):
# One mismatch, second barcode.
barcodes = demultiplex.prepare_barcodes(self.barcodes5, None)
entry1 = FastqEntry(
'@header1',
'TTAATT' + make_sequence(40),
'+',
make_quality_scores(44),
)
entry2 = FastqEntry(
'@header2',
'GGACGAGG' + make_sequence(40),
'+',
make_quality_scores(48),
)
fq_fname = self.create_fq_file([entry1, entry2])
handle = demultiplex._extract(fq_fname,
barcodes,
mismatches=1,
minimum_length=15)
read1, wbrc, randomer1 = next(handle)
self.assertEqual(wbrc, self.barcodes5[0])
self.assertEqual(randomer1, 'TTT')
self.assertEqual(read1.id, entry1.id)
self.assertEqual(read1.seq, entry1.seq[6:])
self.assertEqual(read1.plus, entry1.plus)
self.assertEqual(read1.qual, entry1.qual[6:])
read2, wbrc, randomer2 = next(handle)
self.assertEqual(wbrc, self.barcodes5[2])
self.assertEqual(randomer2, 'GGGG')
self.assertEqual(read2.id, entry2.id)
self.assertEqual(read2.seq, entry2.seq[8:])
self.assertEqual(read2.plus, entry2.plus)
self.assertEqual(read2.qual, entry2.qual[8:])
def test_barcode_size_diff(self):
# One mismatch, second barcode.
barcodes = ['NAAANN', 'NNCCCNN']
adapter = 'CCCCCC'
data = [
'@header1',
'TACATT' + adapter + make_sequence(40, rnd_seed=0),
'+',
make_quality_scores(50, rnd_seed=0) + '!J',
'@header2',
'AACCCTT' + adapter + make_sequence(39, rnd_seed=0),
'+',
make_quality_scores(50, rnd_seed=0) + '!J',
]
fq_fname = get_temp_file_name(extension='fq')
fq_file = iCount.files.fastq.FastqFile(fq_fname, 'wt')
fq_file.write(iCount.files.fastq.FastqEntry(*data[:4]))
fq_file.write(iCount.files.fastq.FastqEntry(*data[4:]))
fq_file.close()
handle = demultiplex._extract(fq_fname, barcodes, mismatches=1)
read1, exp_id1, randomer1 = next(handle)
self.assertEqual(exp_id1, 0)
self.assertEqual(randomer1, 'TTT')
self.assertEqual(read1.id, data[0])
self.assertEqual(read1.seq, data[1][6:])
self.assertEqual(read1.plus, '+')
self.assertEqual(read1.qual, data[3][6:])
read2, exp_id2, randomer2 = next(handle)
self.assertEqual(exp_id2, 1)
self.assertEqual(randomer2, 'AATT')
self.assertEqual(read2.id, data[4])
self.assertEqual(read2.seq, data[5][7:])
self.assertEqual(read2.plus, '+')
self.assertEqual(read2.qual, data[7][7:])
def setUp(self):
warnings.simplefilter("ignore", ResourceWarning)
self.dir = get_temp_dir()
self.adapter = 'AAAAAAAAAA'
self.barcodes5 = [
'NNAAAN',
'NGGGN',
'NGGGN',
]
self.barcodes3 = [
'.',
'NNGGG',
'NCCC',
]
# Header: early version Illumina header
# Barcodes: exact match to the barcode set #1
self.entry1 = FastqEntry(
'@header1/1',
'GGAAAG' + make_sequence(40) + self.adapter,
'+',
make_quality_scores(56),
)
# Header: contains id and description
# Barcodes: one mismatch on 5' end for barcode set #2
self.entry2 = FastqEntry(
'@header2 blah',
'AGGTA' + make_sequence(40) + 'AAGGG' + self.adapter,
'+',
make_quality_scores(60),
)
# Header: simple header
# Barcodes: one mismatch on 3' end for barcode set #3
self.entry3 = FastqEntry(
'@header3',
'TGGGT' + make_sequence(40) + 'TACC' + self.adapter,
'+',
make_quality_scores(59),
)
self.fq_fname = get_temp_file_name(extension='fq')
self.fq_file = iCount.files.fastq.FastqFile(self.fq_fname, 'wt')
for entry in [self.entry1, self.entry2, self.entry3]:
self.fq_file.write(entry)
self.fq_file.close()
def test_barcode3(self):
# One mismatch, second barcode.
barcodes = demultiplex.prepare_barcodes(
self.barcodes5 + [self.barcodes5[2]], self.barcodes3)
barcodes = barcodes[self.barcodes5[2]]['barcodes3']
entry1 = FastqEntry(
'@header1',
make_sequence(40) + 'TTGGG',
'+',
make_quality_scores(45),
)
entry2 = FastqEntry(
'@header2',
make_sequence(40) + 'TCAAA',
'+',
make_quality_scores(45),
)
fq_fname = self.create_fq_file([entry1, entry2])
handle = demultiplex._extract(fq_fname,
barcodes,
mismatches=1,
minimum_length=15)
read1, wbrc, randomer1 = next(handle)
self.assertEqual(wbrc, self.barcodes3[2])
self.assertEqual(randomer1, 'TT')
self.assertEqual(read1.id, entry1.id)
self.assertEqual(read1.seq, entry1.seq[:-5])
self.assertEqual(read1.plus, entry1.plus)
self.assertEqual(read1.qual, entry1.qual[:-5])
read2, wbrc, randomer2 = next(handle)
self.assertEqual(wbrc, self.barcodes3[3])
self.assertEqual(randomer2, 'TC')
self.assertEqual(read2.id, entry2.id)
self.assertEqual(read2.seq, entry2.seq[:-5])
self.assertEqual(read2.plus, entry2.plus)
self.assertEqual(read2.qual, entry2.qual[:-5])
def test_extract_mismatch(self):
# To many mismatches
barcodes = demultiplex.prepare_barcodes(self.barcodes5, None)
entry = FastqEntry(
'@header1',
'TTTTTT' + make_sequence(40),
'+',
make_quality_scores(46),
)
fq_fname = self.create_fq_file([entry])
for read, wbrc, randomer in demultiplex._extract(fq_fname,
barcodes,
mismatches=0,
minimum_length=15):
self.assertEqual(wbrc, 'nomatch')
self.assertEqual(randomer, '')
self.assertEqual(read.id, entry.id)
self.assertEqual(read.seq, entry.seq)
self.assertEqual(read.plus, entry.plus)
self.assertEqual(read.qual, entry.qual)
def test_extract_ok_2(self):
# One mismatch, second barcode.
barcodes = demultiplex.prepare_barcodes(self.barcodes5, None)
entry = FastqEntry(
'@header1',
'TTAGTT' + make_sequence(40),
'+',
make_quality_scores(46),
)
fq_fname = self.create_fq_file([entry])
for read, wbrc, randomer in demultiplex._extract(fq_fname,
barcodes,
mismatches=1,
minimum_length=15):
self.assertEqual(wbrc, self.barcodes5[1])
self.assertEqual(randomer, 'TTT')
self.assertEqual(read.id, entry.id)
self.assertEqual(read.seq, entry.seq[6:])
self.assertEqual(read.plus, entry.plus)
self.assertEqual(read.qual, entry.qual[6:])
def test_extract_mismatch(self):
# To many mismatches
barcodes = ['NNAAAN', 'NNCCTN', 'NNACGN']
adapter = 'CCCCCC'
data = [
'@header1', 'TTACTT' + adapter + make_sequence(40, rnd_seed=0),
'+',
make_quality_scores(50, rnd_seed=0) + '!J'
]
fq_fname = get_temp_file_name(extension='fq')
fq_file = iCount.files.fastq.FastqFile(fq_fname, 'wt')
fq_file.write(iCount.files.fastq.FastqEntry(*data))
fq_file.close()
for read, exp_id, randomer in demultiplex._extract(fq_fname,
barcodes,
mismatches=0):
self.assertEqual(exp_id, -1)
self.assertEqual(randomer, '')
self.assertEqual(read.id, data[0])
self.assertEqual(read.seq, data[1])
self.assertEqual(read.plus, '+')
self.assertEqual(read.qual, data[3])
def setUp(self):
warnings.simplefilter("ignore", (ResourceWarning, ImportWarning))
# Make fai and fasta file:
self.fai_data = {'1': 1000}
self.fai = make_file_from_list(list(self.fai_data.items()), bedtool=False, extension='fai')
random.seed(0) # pylint:disable=no-member
random_seeds = random.randint(10**5, size=len(self.fai)) # pylint:disable=no-member
self.seqs = {name: make_sequence(size, rnd_seed=rseed) for (name, size), rseed in
zip(self.fai_data.items(), random_seeds)}
self.fasta = make_fasta_file(sequences=self.seqs.values(), headers=self.seqs.keys())
# Make annotation:
self.gtf = make_file_from_list(extension='gtf', data=[
# Gene #1, positive strand:
['1', '.', 'gene', '100', '499', '.', '+', '.', attrs(gid='G1')],
# Transcript #1
['1', '.', 'transcript', '100', '249', '.', '+', '.', attrs(gid='G1', tid='T1')],
['1', '.', 'exon', '100', '149', '.', '+', '.', attrs(gid='G1', tid='T1', exn=1)],
['1', '.', 'exon', '200', '229', '.', '+', '.', attrs(gid='G1', tid='T1', exn=2)],
['1', '.', 'CDS', '200', '229', '.', '+', '.', attrs(gid='G1', tid='T1', exn=2)],
['1', '.', 'exon', '240', '249', '.', '+', '.', attrs(gid='G1', tid='T1', exn=3)],
# Transcript #2
['1', '.', 'transcript', '240', '499', '.', '+', '.', attrs(gid='G1', tid='T2')],
['1', '.', 'exon', '240', '299', '.', '+', '.', attrs(gid='G1', tid='T2', exn=1)],
['1', '.', 'CDS', '240', '299', '.', '+', '.', attrs(gid='G1', tid='T2', exn=1)],
['1', '.', 'exon', '400', '499', '.', '+', '.', attrs(gid='G1', tid='T2', exn=2)],
['1', '.', 'CDS', '400', '499', '.', '+', '.', attrs(gid='G1', tid='T2', exn=2)],
# Gene #2, positive strand:
['1', '.', 'gene', '600', '799', '.', '+', '.', attrs(gid='G2')],
# Transcript #3
['1', '.', 'transcript', '600', '799', '.', '+', '.', attrs(gid='G2', tid='T3')],
['1', '.', 'exon', '600', '649', '.', '+', '.', attrs(gid='G2', tid='T3', exn=1)],
['1', '.', 'CDS', '600', '649', '.', '+', '.', attrs(gid='G2', tid='T3', exn=1)],
['1', '.', 'exon', '750', '799', '.', '+', '.', attrs(gid='G2', tid='T3', exn=2)],
['1', '.', 'CDS', '750', '799', '.', '+', '.', attrs(gid='G2', tid='T3', exn=2)],
# Gene #3, negative strand:
['1', '.', 'gene', '800', '899', '.', '-', '.', 'gene_id "G3";'],
# Transcript #4
['1', '.', 'transcript', '800', '899', '.', '-', '.', attrs(gid='G3', tid='T4')],
['1', '.', 'exon', '800', '899', '.', '-', '.', attrs(gid='G3', tid='T4', exn=1)],
['1', '.', 'CDS', '800', '899', '.', '-', '.', attrs(gid='G3', tid='T4', exn=1)],
])
# Define positions of cross-links:
self.read_data = [
# [name, chromosome, sequence_len, read_start, strand]
('name01:rbc:CCCC', '1', 50, 80, '+'),
('name02:rbc:CCCC', '1', 50, 140, '+'),
('name03:rbc:AAAA', '1', 50, 142, '+'),
('name04:rbc:CCCC', '1', 50, 142, '+'),
('name05:rbc:CCCC', '1', 30, 160, '+'),
('name06:rbc:CCCC', '1', 30, 163, '+'),
('name07:rbc:GGGG', '1', 30, 163, '+'),
('name08:rbc:CCCC', '1', 20, 205, '+'),
('name09:rbc:CCCC', '1', 50, 235, '+'),
('name10:rbc:CCCC', '1', 50, 480, '+'),
('name11:rbc:CCCC', '1', 30, 530, '+'),
('name12:rbc:CCCC', '1', 30, 610, '+'),
('name13:rbc:CCCC', '1', 30, 549, '-'),
]
# Make reads from genome, pointing to specific pre-detemrined cross-link positions:
random_seeds = random.randint(10**5, size=len(self.read_data)) # pylint:disable=no-member
self.reads = get_temp_file_name(extension='fastq')
with open(self.reads, 'wt') as ofile:
for (name, chrom, length, xlink, strand), rseed in zip(self.read_data, random_seeds):
seq = self.seqs[chrom][xlink: xlink + length]
if strand == '-':
seq = self.seqs[chrom][xlink - length: xlink]
compl_bases = {'A': 'T', 'C': 'G', 'G': 'C', 'T': 'A'}
seq = ''.join([compl_bases[nuc] for nuc in seq[::-1]])
ofile.write('@' + name + '\n')
ofile.write(seq + '\n')
ofile.write('+' + '\n')
ofile.write(
make_quality_scores(len(seq), min_chr=65, max_chr=73, rnd_seed=rseed) + '\n')
# Make a "strange" read:
# pylint: disable=undefined-loop-variable
seq = self.seqs[chrom][250: 295] + self.seqs[chrom][310: 380]
ofile.write('@' + 'name_strange:rbc:GGGG' + '\n')
ofile.write(seq + '\n')
ofile.write('+' + '\n')
ofile.write(
make_quality_scores(len(seq), min_chr=70, max_chr=73, rnd_seed=rseed) + '\n')
def test_run_ok_no_adapter(self):
barcodes = ['NNAAAN', 'NNACTN']
data = [
[
'@header/1', 'GGAAAG' + make_sequence(40, rnd_seed=99), '+',
make_quality_scores(50, min_chr=65, max_chr=73, rnd_seed=76) +
'!J'
],
[
'@header2 blah', 'TTCCTT' + make_sequence(40, rnd_seed=47),
'+',
make_quality_scores(50, min_chr=65, max_chr=73, rnd_seed=71) +
'!J'
],
[
'@header3', 'TTGGGT' + make_sequence(40, rnd_seed=13), '+',
make_quality_scores(50, min_chr=65, max_chr=73, rnd_seed=12) +
'!J'
],
]
fq_fname = get_temp_file_name(extension='fq')
fq_file = iCount.files.fastq.FastqFile(fq_fname, 'wt')
for line in data:
fq_file.write(iCount.files.fastq.FastqEntry(*line))
fq_file.close()
demultiplex.run(fq_fname,
None,
barcodes,
mismatches=1,
out_dir=self.dir)
fq1_list = make_list_from_file('{}/demux_{}.fastq.gz'.format(
self.dir, barcodes[0]))
expected1 = [
['@header:rbc:GGG/1'],
[data[0][1][6:]],
['+'],
[data[0][3][6:]],
]
self.assertEqual(fq1_list, expected1)
fq2_list = make_list_from_file('{}/demux_{}.fastq.gz'.format(
self.dir, barcodes[1]))
expected2 = [
['@header2:rbc:TTT'],
[data[1][1][6:]],
['+'],
[data[1][3][6:]],
]
self.assertEqual(fq2_list, expected2)
fq3_list = make_list_from_file('{}/demux_{}.fastq.gz'.format(
self.dir, 'nomatch'))
expected3 = [
['@header3'],
[data[2][1]],
['+'],
[data[2][3]],
]
self.assertEqual(fq3_list, expected3)
