def index(args):
if not os.path.exists(args.indexdir):
print "Index_dir %s does not exist!" % (args.indexdir)
sys.exit(1)
fasta_dir = args.fastadir
index_dir = os.path.join(args.indexdir, args.indexname)
g = GenomeIndex()
g.create_index(fasta_dir, index_dir)
# Create genome FASTA file for use with bedtools
with open(os.path.join(index_dir, "genome.fa"), 'w') as out:
for f in find_by_ext(fasta_dir, FASTA_EXT):
for line in open(f):
out.write(line)
test_chr = g.get_chromosomes()[0]
tmp = NamedTemporaryFile()
tmp.write("{}\t1\t2\n".format(test_chr))
tmp.flush()
b = pybedtools.BedTool(tmp.name)
try:
b.nucleotide_content(fi=os.path.join(index_dir, "genome.fa"))
except pybedtools.helpers.BEDToolsError as e:
if str(e).find("generating") == -1:
raise
def create_bedtools_fa(index_dir, fasta_dir):
g = GenomeIndex(index_dir)
genome_fa = os.path.join(index_dir, "genome.fa")
# Create genome FASTA file for use with bedtools
with open(genome_fa, 'w') as out:
for fname in find_by_ext(fasta_dir, FASTA_EXT):
with open(fname) as f:
for line in f:
out.write(line)
# Delete old bedtools index if it exists, otherwise bedtools will
# give an error.
if os.path.exists(genome_fa + ".fai"):
os.unlink(genome_fa + ".fai")
test_chr = g.get_chromosomes()[0]
tmp = NamedTemporaryFile(mode="w")
tmp.write("{}\t1\t2\n".format(test_chr))
tmp.flush()
b = pybedtools.BedTool(tmp.name)
try:
# pylint: disable=unexpected-keyword-arg
b.nucleotide_content(fi=genome_fa)
except pybedtools.helpers.BEDToolsError as e:
if str(e).find("generating") == -1:
raise
def create_bedtools_fa(index_dir, fasta_dir):
g = GenomeIndex(index_dir)
# Create genome FASTA file for use with bedtools
with open(os.path.join(index_dir, "genome.fa"), 'w') as out:
for f in find_by_ext(fasta_dir, FASTA_EXT):
for line in open(f):
out.write(line)
test_chr = g.get_chromosomes()[0]
tmp = NamedTemporaryFile()
tmp.write("{}\t1\t2\n".format(test_chr))
tmp.flush()
b = pybedtools.BedTool(tmp.name)
try:
b.nucleotide_content(fi=os.path.join(index_dir, "genome.fa"))
except pybedtools.helpers.BEDToolsError as e:
if str(e).find("generating") == -1:
raise
def create_bedtools_fa(index_dir, fasta_dir):
g = GenomeIndex(index_dir)
# Create genome FASTA file for use with bedtools
with open(os.path.join(index_dir, "genome.fa"), 'w') as out:
for f in find_by_ext(fasta_dir, FASTA_EXT):
for line in open(f):
out.write(line)
test_chr = g.get_chromosomes()[0]
tmp = NamedTemporaryFile()
tmp.write("{}\t1\t2\n".format(test_chr))
tmp.flush()
b = pybedtools.BedTool(tmp.name)
try:
b.nucleotide_content(fi=os.path.join(index_dir, "genome.fa"))
except pybedtools.helpers.BEDToolsError as e:
if str(e).find("generating") == -1:
raise
class GenomeIndex:
""" Index fasta-formatted files for faster retrieval of sequences
Typical use:
# Make index
g = GenomeIndex()
g.create_index("/usr/share/genomes/hg18", "/usr/share/genome_index/hg18")
# Retrieve sequence
g = GenomeIndex("/usr/share/genome_index/hg18")
seq = g.get_sequence("chr17", "7520037", "7531588")
# Batch bed-file to fasta-file
track2fasta("/usr/share/genome_index/hg18", "p53_targets.bed", "p53_targets.fa")
"""
def __init__(self, index_dir=None):
""" Initialize GenomeIndex with index_dir as optional argument"""
self.param_file = "index.params"
self.size_file = "genome.size"
self.index_dir = index_dir
self.fasta_dir = None
self.size = {}
self.fasta_file = {}
self.index_file = {}
self.line_size = {}
self.pack_char = "L"
if self.index_dir:
if os.path.exists(os.path.join(self.index_dir, self.param_file)):
self._read_index_file()
def _check_dir(self, dir):
""" Check if dir exists, if not: give warning and die"""
if not os.path.exists(dir):
print "Directory %s does not exist!" % dir
sys.exit(1)
def _make_index(self, fasta, index):
""" Index a single, one-sequence fasta-file"""
out = open(index, "w")
f = open(fasta)
# Skip first line of fasta-file
line = f.readline()
offset = f.tell()
line = f.readline()
while line:
out.write(pack(self.pack_char, offset))
offset = f.tell()
line = f.readline()
f.close()
out.close()
def create_index(self, fasta_dir=None, index_dir=None):
"""Index all fasta-files in fasta_dir (one sequence per file!) and
store the results in index_dir"""
# Use default directories if they are not supplied
if not fasta_dir:
fasta_dir = self.fasta_dir
if not index_dir:
index_dir = self.index_dir
# Can't continue if we still don't have an index_dir or fasta_dir
if not fasta_dir:
print "fasta_dir not defined!"
sys.exit(1)
if not index_dir:
print "index_dir not defined!"
sys.exit(1)
index_dir = os.path.abspath(index_dir)
fasta_dir = os.path.abspath(fasta_dir)
self.index_dir = index_dir
# Prepare index directory
if not os.path.exists(index_dir):
try:
os.mkdir(index_dir)
except OSError, e:
if e.args[0] == 13:
sys.stderr.write(
"No permission to create index directory. Superuser access needed?\n"
)
sys.exit()
else:
sys.stderr.write(e)
# Directories need to exist
self._check_dir(fasta_dir)
self._check_dir(index_dir)
# Get all fasta-files
fastafiles = find_by_ext(fasta_dir, FASTA_EXT)
if not (fastafiles):
raise IOError, \
"No fastafiles found in {} with extension in {}".format(
fasta_dir, ",".join(FASTA_EXT))
# param_file will hold all the information about the location of the fasta-files, indeces and
# length of the sequences
param_file = os.path.join(index_dir, self.param_file)
size_file = os.path.join(index_dir, self.size_file)
try:
out = open(param_file, "w")
except IOError, e:
if e.args[0] == 13:
sys.stderr.write(
"No permission to create files in index directory. Superuser access needed?\n"
)
sys.exit()
else:
sys.stderr.write(e)
def create_index(self,fasta_dir=None, index_dir=None):
"""Index all fasta-files in fasta_dir (one sequence per file!) and
store the results in index_dir"""
# Use default directories if they are not supplied
if not fasta_dir:
fasta_dir = self.fasta_dir
if not index_dir:
index_dir = self.index_dir
# Can't continue if we still don't have an index_dir or fasta_dir
if not fasta_dir:
print("fasta_dir not defined!")
sys.exit(1)
if not index_dir:
print("index_dir not defined!")
sys.exit(1)
index_dir = os.path.abspath(index_dir)
fasta_dir = os.path.abspath(fasta_dir)
self.index_dir = index_dir
# Prepare index directory
if not os.path.exists(index_dir):
try:
os.mkdir(index_dir)
except OSError as e:
if e.args[0] == 13:
sys.stderr.write("No permission to create index directory. Superuser access needed?\n")
sys.exit()
else:
sys.stderr.write(e)
# Directories need to exist
self._check_dir(fasta_dir)
self._check_dir(index_dir)
# Get all fasta-files
fastafiles = find_by_ext(fasta_dir, FASTA_EXT)
if not(fastafiles):
msg = "No fastafiles found in {} with extension in {}".format(
fasta_dir, ",".join(FASTA_EXT))
raise IOError(msg)
# param_file will hold all the information about the location of the fasta-files, indeces and
# length of the sequences
param_file = os.path.join(index_dir, self.param_file)
size_file = os.path.join(index_dir, self.size_file)
try:
out = open(param_file, "w")
except IOError as e:
if e.args[0] == 13:
sys.stderr.write("No permission to create files in index directory. Superuser access needed?\n")
sys.exit()
else:
sys.stderr.write(e)
s_out = open(size_file, "w")
for fasta_file in fastafiles:
#sys.stderr.write("Indexing %s\n" % fasta_file)
f = open(fasta_file)
line = f.readline()
if not line.startswith(">"):
sys.stderr.write("%s is not a valid FASTA file, expected > at first line\n" % fasta_file)
sys.exit()
seqname = line.strip().replace(">", "")
line = f.readline()
line_size = len(line.strip())
total_size = 0
while line:
line = line.strip()
if line.startswith(">"):
sys.stderr.write("Sorry, can only index genomes with "
"one sequence per FASTA file\n%s contains multiple "
"sequences\n" % fasta_file)
sys.exit()
total_size += len(line)
line = f.readline()
index_file = os.path.join(index_dir, "%s.index" % seqname)
out.write("{}\t{}\t{}\t{}\t{}\n".format(
seqname, fasta_file, index_file, line_size, total_size))
s_out.write("{}\t{}\n".format(seqname, total_size))
self._make_index(fasta_file, index_file)
f.close()
out.close()
s_out.close()
# Read the index we just made so we can immediately use it
self._read_index_file()