def test_format_contigs_denovo():
# test with a custom fasta
filename = sequana_data("test_fasta.fasta")
contigs = FastA(filename)
with TempFile(suffix='.fasta') as fh:
contigs.format_contigs_denovo(fh.name)
contigs.names
contigs.lengths
contigs.comments
def __init__(self, directory=".", prefix=""):
self.prefix = prefix
self.directory = directory
self.sample_name = "undefined"
# low quality isoforms
filename = "all.polished_lq.fastq"
self.lq_isoforms = self.get_file(filename)
if self.lq_isoforms:
logger.info("Reading {}".format(filename))
self.lq_sequence = FastQ(self.lq_isoforms)
# high quality isoforms
filename = "all.polished_hq.fastq"
self.hq_isoforms = self.get_file(filename)
if self.hq_isoforms:
logger.info("Reading {}".format(filename))
self.hq_sequence = FastQ(self.hq_isoforms)
# General info
filename = "file.csv"
self.csv = self.get_file(filename)
if self.csv:
logger.info("Reading {}".format(filename))
self.data = pd.read_csv(self.csv)
# CCS fasta sequence
#self.ccs = self.get_file("-ccs.tar.gz")
filename = "ccs.fasta"
self.ccs = self.get_file(filename, noprefix=True)
if self.ccs:
logger.info("Reading {}".format(filename))
self.ccs = FastA(self.ccs)
def find_motif_fasta(self, filename, motif, window=200,
local_threshold=None, global_threshold=None):
from sequana import FastA
data = FastA(filename)
N = len(data)
from easydev import Progress
pb = Progress(N)
df = {
"query_name": [],
"hit": [],
"length": [],
"start": [],
"end": []
}
for i, item in enumerate(data):
X1, S = self.find_motif_from_sequence(item.sequence, motif,
window=window, local_threshold=local_threshold
)
if S >= self.global_threshold:
df['query_name'].append(item.name)
df['start'].append(0)
df['end'].append(len(item.sequence))
df['length'].append(len(item.sequence))
df['hit'].append(S)
pb.animate(i+1)
df = pd.DataFrame(df)
return df
def main(args=None):
if args is None:
args = sys.argv[:]
user_options = Options(prog="sequana")
# If --help or no options provided, show the help
if len(args) == 1:
user_options.parse_args(["prog", "--help"])
else:
options = user_options.parse_args(args[1:])
reference = options.reference
if options.file1 and options.file2:
fastq = "%s %s" % (options.file1, options.file2)
elif options.file1 and not options.file2:
fastq = "%s" % (options.file1)
elif options.file1 is None:
raise ValueError("--file1 must be used")
from sequana import FastQ
from sequana import FastA
S = 0
for this in FastQ(options.file1):
S += len(this['sequence'])
if options.file2:
for this in FastQ(options.file2):
S += len(this['sequence'])
ref = FastA(options.reference)
coverage = float(S) / len(ref.sequences[0])
print('Theoretical Depth of Coverage : %s' % coverage)
params = {"reference": reference, "fastq": fastq, "thread": options.thread}
# indexing
shellcmd("bwa index %(reference)s " % params)
cmd = "samtools faidx %(reference)s " % params
# mapping
cmd = "bwa mem -M " # mark shorter split read as secondary; -M is not compulsary but recommended
if options.pacbio:
cmd += "-x pacbio "
cmd += r" -t %(thread)s -R @RG\\tID:1\\tSM:1\\tPL:illumina -T 30 %(reference)s %(fastq)s "
# Samtools options:
# S:ignore input format
# h:include header
# b:bam output
if options.sambamba is False:
cmd += "| samtools view -Sbh | "
# sorting BAM
cmd += "samtools sort -@ %(thread)s -o %(reference)s.sorted.bam -"
shellcmd(cmd % params)
else:
# FIXME use sambamba for the view as well
cmd += "| samtools view -Sbu - | sambamba sort /dev/stdin -o %(reference)s.sorted.bam -t %(thread)s --tmpdir=./tmp " % params
shellcmd(cmd % params)
def test_fasta_fwd_rev_to_columns():
a1 = sequana_data("adapters_PCRFree_fwd.fa")
a2 = sequana_data("adapters_PCRFree_rev.fa")
f1 = FastA(a1)
f2 = FastA(a2)
assert f1 == f1
assert f1 != f2
assert len(f1) == 49
assert len(f2) == 49
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, a2, fh.name)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, None, output_filename=fh.name)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, a2)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, None)
def test_fasta_fwd_rev_to_columns():
a1 = sequana_data("NEXTFlex48_DNA_fwd.fa")
a2 = sequana_data("NEXTFlex48_DNA_rev.fa")
a3 = sequana_data("NEXTFlex48_DNA_revcomp.fa")
f1 = FastA(a1)
f2 = FastA(a2)
f3 = FastA(a3)
assert f1 == f1
assert f1 != f2
assert len(f1) == 49
assert len(f2) == 49
assert len(f3) == 49
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, a2, fh.name)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, None, output_filename=fh.name)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, a2)
with TempFile() as fh:
adapters.fasta_fwd_rev_to_columns(a1, None)
def add_locus_in_fasta(self, fasta, output_file):
""" Add locus of annotation file in description line of fasta file. If
fasta file and genbank file do not have the same names.
:param str fasta: input fasta file where you want to add locus.
:param str output_file: output file.
FIXME: fasta is already known if provided in the init
"""
fasta_record = FastA(fasta)
ids_list = self._get_seq_ids()
# check if both files have same number of contigs
if len(fasta_record) != len(ids_list):
print("fasta and annotation files don't have the same number of "
"contigs. Found {} and {}".format(len(fasta_record),
len(ids_list)))
sys.exit(1)
# check if directory exist
output_dir = os.path.dirname(output_file)
try:
if not os.path.exists(output_dir):
os.makedirs(output_dir)
except FileNotFoundError:
pass
if sorted(fasta_record.names) == sorted(ids_list):
logger.info("Files have same sequence id.")
if os.path.isfile(output_file):
os.remove(output_file)
os.symlink(os.path.realpath(fasta), output_file)
return
else:
logger.info(
"fasta and GFF seem to have different IDs. Creating a"
"new coherent fasta file assuming the chromsome names appear "
"in the same order in the fasta and gff")
with open(output_file, "w") as fp:
# write fasta with seqid of annotation file
for n in range(len(fasta_record)):
seq_id = ">{0} {1}\n".format(ids_list[n],
fasta_record.names[n])
seq = fasta_record.sequences[n]
sequence = "\n".join([
seq[i:min(i + 80, len(seq))]
for i in range(0, len(seq), 80)
]) + "\n"
contigs = seq_id + sequence
fp.write(contigs)
def bar_plot_contigs_length(self):
# show length of N contigs as compare to length of the reference
fref = FastA(self.reference)
Nref = len(fref.sequences)
N = len(self.fasta)
pylab.clf()
pylab.bar(range(0, N, int(pylab.ceil(N / Nref))),
sorted(fref.lengths),
width=Nref / 1.1,
label="Plasmodium chromosomes")
pylab.bar(range(0, N),
sorted(self.fasta.lengths),
width=1,
label="canu {} contigs".format(N))
pylab.legend()
def __init__(self, filename, reference=None, bamfile=None, mode="canu"):
"""
minimap2 -x map-pb reference filename -a > temp.sam
bioconvert sam2bam temp.sam temp.bam
"""
self.filename = filename
self.fasta = FastA(filename)
self.mode = mode
self._df = None
if bamfile:
self.bam = BAM(bamfile)
else:
self.bam = None
self.reference = reference
def add_locus_in_fasta(self, fasta, output_file):
""" Add locus of annotation file in description line of fasta file. If
fasta file and genbank file do not have the same names.
:param str fasta: input fasta file where you want to add locus.
:param str output_file: output file.
"""
fasta_record = FastA(fasta)
ids_list = self._get_seq_ids()
# check if both files have same number of contigs
if len(fasta_record) != len(ids_list):
print("fasta and annotation files don't have the same number of "
"contigs.")
sys.exit(1)
# check if directory exist
output_dir = os.path.dirname(output_file)
try:
if not os.path.exists(output_dir):
os.makedirs(output_dir)
except FileNotFoundError:
pass
if fasta_record.names[0] == ids_list[0]:
print("Files have same sequence id.")
if os.path.isfile(output_file):
os.remove(output_file)
os.symlink(os.path.realpath(fasta), output_file)
return
with open(output_file, "w") as fp:
# write fasta with seqid of annotation file
for n in range(len(fasta_record)):
seq_id = ">{0} {1}\n".format(ids_list[n],
fasta_record.names[n])
seq = fasta_record.sequences[n]
sequence = "\n".join([
seq[i:min(i + 80, len(seq))]
for i in range(0, len(seq), 80)
]) + "\n"
contigs = seq_id + sequence
fp.write(contigs)
def __init__(self, directory=".", prefix="job-*"):
self.prefix = prefix
self.directory = directory
# low quality isoforms
self.lq_isoforms = self.get_file("lq_isoforms.fastq")
if self.lq_isoforms:
self.lq_sequence = FastQ(self.lq_isoforms)
# high quality isoforms
self.hq_isoforms = self.get_file("hq_isoforms.fastq")
if self.hq_isoforms:
self.hq_sequence = FastQ(self.hq_isoforms)
# General info
self.csv = self.get_file("-file.csv")
if self.csv:
self.data = pd.read_csv(self.csv)
# CCS fasta sequence
#self.ccs = self.get_file("-ccs.tar.gz")
self.ccs = self.get_file("ccs.fasta", noprefix=True)
if self.ccs:
self.ccs = FastA(self.ccs)
def get_fasta_stats(filename, sample=1e16):
from sequana import FastA
ff = FastA(filename)
stats = ff.get_stats()
return stats
def __init__(self, filename, shift=4):
from sequana import FastA
self.SIRV = FastA(filename)
self.shift = 5
def __init__(self, filename):
self.filename = filename
self.fasta = FastA(filename)
def test_others():
filename = sequana_data("test_fasta.fasta")
ff = FastA(filename)
assert len(ff) == 16
assert len(ff.comments) == 16
assert len(ff.names) == 16
assert len(ff.sequences) == 16
assert is_fasta(filename) == True
ff.get_lengths_as_dict()
with TempFile(suffix='.fasta') as fh:
ff.select_random_reads(4, output_filename=fh.name)
ff.select_random_reads([1, 2, 3], output_filename=fh.name)
ff.select_random_reads({1, 2, 3}, output_filename=fh.name)
assert ff.get_stats()['N'] == 16
assert ff.get_stats()['mean_length'] > 454
with TempFile(suffix='.fasta') as fh:
ff.reverse_and_save(fh.name)
ff.to_fasta(fh.name)
ff.to_igv_chrom_size(fh.name)
def test_fasta_filtering():
filename = sequana_data("test_fasta_filtering.fa")
ff = FastA(filename)
with TempFile(suffix='.fasta') as fh:
ff.to_fasta(fh.name)
ff.save_ctg_to_fasta("A", fh.name)
with TempFile(suffix='.fasta') as fh:
ff.filter(fh.name, names_to_exclude=["A", "B"])
reader = FastA(fh.name)
assert set(reader.names) == set(["C", "D"])
ff = FastA(filename)
with TempFile(suffix='.fasta') as fh:
ff.filter(fh.name, names_to_keep=[
"A",
])
reader = FastA(fh.name)
assert set(reader.names) == set(['A'])