def discosnp_install():
print(_os.listdir('.'))
_subprocess.call(['bash','./compile_discoSnp++.sh'])
# ensure VCF generation works if system default python is version 3
# VCF code is python 2 only
fixed = open('run_VCF_creator.sh').read().replace('python ','python2 ')
open('run_VCF_creator.sh','w').write(fixed)
def discosnp_install():
print(_os.listdir('.'))
_subprocess.call(['bash', './compile_discoSnp++.sh'])
# ensure VCF generation works if system default python is version 3
# VCF code is python 2 only
fixed = open('run_VCF_creator.sh').read().replace('python ', 'python2 ')
open('run_VCF_creator.sh', 'w').write(fixed)
def pysam_install():
_subprocess.call([_sys.executable, 'setup.py', 'build'])
for f in _os.listdir('build/lib.linux-x86_64-2.7/pysam'):
if f[-3:] == '.so':
#print(f)
_shutil.copy('build/lib.linux-x86_64-2.7/pysam/{}'.format(f), 'pysam/{}'.format(f))
try:
_shutil.rmtree('../pysam')
except OSError:
pass
_os.rename('pysam', '../pysam')
def pysam_install():
_subprocess.call([_sys.executable, 'setup.py', 'build'])
for f in _os.listdir('build/lib.linux-x86_64-2.7/pysam'):
if f[-3:] == '.so':
#print(f)
_shutil.copy('build/lib.linux-x86_64-2.7/pysam/{}'.format(f),
'pysam/{}'.format(f))
try:
_shutil.rmtree('../pysam')
except OSError:
pass
_os.rename('pysam', '../pysam')
def prep_simple_make(path=False, configure=False, alt_command=False):
if path:
_os.chdir(_os.path.sep.join(path))
if configure:
_subprocess.call(['./configure'])
if alt_command:
_subprocess.call([alt_command])
else:
_subprocess.call(['make'])
if path:
_os.chdir(_os.path.sep.join([_os.path.pardir] * len(path)))
def prep_simple_make(path = False, configure = False, alt_command = False):
if path:
_os.chdir(_os.path.sep.join(path))
if configure:
_subprocess.call(['./configure'])
if alt_command:
_subprocess.call([alt_command])
else:
_subprocess.call(['make'])
if path:
_os.chdir(_os.path.sep.join([_os.path.pardir]*len(path)))
def get_git(name, description, source, url, commit, checksum, destination,
preparation, checker):
'''
Get a dependency from git
'''
if _os.path.realpath(_os.path.curdir) != destination:
try:
_os.chdir(destination)
except OSError:
_os.makedirs(destination)
_os.chdir(destination)
try:
# clear any previous verions
_shutil.rmtree(url.split('/')[-1].replace('.git', ''))
except OSError:
pass
git_server = url.replace('https://', '').replace('http://',
'').split('/')[0]
print('Downloading {} via git from {} . . .'.format(name, git_server))
_subprocess.call(['git', 'clone', url])
_os.chdir(url.split('/')[-1].replace('.git', ''))
_subprocess.call(['git', 'checkout', commit])
# if repo uses git submodules, those will be set to the correct revisions for this commit
# else will do nothing
_subprocess.call(['git', 'submodule', 'update', '--init'])
working_dir = _os.path.sep.join(
[destination, url.split('/')[-1].replace('.git', '')])
if preparation is not None:
for do_this in preparation:
if isinstance(do_this['arguments'], dict):
do_this['function'](**do_this['arguments'])
else:
do_this['function'](*do_this['arguments'])
# restore position in path if a prepare changed it
if working_dir != _os.path.realpath(_os.path.curdir):
_os.chdir(working_dir)
_os.chdir(_os.path.pardir)
_os.chdir(_os.path.pardir)
def get_git(name, description, source, url, commit, checksum, destination, preparation, checker):
'''
Get a dependency from git
'''
if _os.path.realpath(_os.path.curdir) != destination:
try:
_os.chdir(destination)
except OSError:
_os.makedirs(destination)
_os.chdir(destination)
try:
# clear any previous verions
_shutil.rmtree(url.split('/')[-1].replace('.git',''))
except OSError:
pass
git_server = url.replace('https://','').replace('http://','').split('/')[0]
print('Downloading {} via git from {} . . .'.format(name, git_server))
_subprocess.call(['git', 'clone', url])
_os.chdir(url.split('/')[-1].replace('.git',''))
_subprocess.call(['git', 'checkout', commit])
# if repo uses git submodules, those will be set to the correct revisions for this commit
# else will do nothing
_subprocess.call(['git', 'submodule', 'update', '--init'])
working_dir = _os.path.sep.join([destination,url.split('/')[-1].replace('.git','')])
if preparation is not None:
for do_this in preparation:
if isinstance(do_this['arguments'], dict):
do_this['function'](**do_this['arguments'])
else:
do_this['function'](*do_this['arguments'])
# restore position in path if a prepare changed it
if working_dir != _os.path.realpath(_os.path.curdir):
_os.chdir(working_dir)
_os.chdir(_os.path.pardir)
_os.chdir(_os.path.pardir)
def prep_python_install(extras):
print('Installing via setup.py . . .')
# could try here?
_subprocess.call([_sys.executable, 'setup.py', 'install'] + extras)
def IndelRealignGATK(self,
jar = ['external_programs', 'GenomeAnalysisTK', 'GenomeAnalysisTK.jar'],
picard_jar = False,
samtools_exe = False,
use_java = 'java',
force = False,
mem_num_gigs = 2,
max_cpus = -1):
# GATK is manually downloaded by user and placed in folder of their choice
jar = _os.path.sep.join(jar)
if not picard_jar:
picard_jar = _get_jar_path('picard')
if not samtools_exe:
samtools_exe = _get_exe_path('samtools')
genome_fna = 'genome_sequences/{}.fna'.format(self.genome_id)
e1 = 'Could not find "paths_to_BAMs_dd_si" attribute. Before starting GATK analysis, read alignments must have duplicates removed. Please run: .toBAMS(), .removeDuplicates(), .sortIndexBAMs() methods on this SAMs instance, or --deduplicate if using baga_cli.py.'
assert hasattr(self, 'paths_to_BAMs_dd_si'), e1
e2 = 'Could not find %s. Please ensure file exists'
for BAM in self.paths_to_BAMs_dd_si:
assert _os.path.exists(BAM), e2 % BAM
if not _os.path.exists(genome_fna[:-4] + '.dict'):
print('Creating sequence dictionary for %s' % genome_fna)
_subprocess.call([use_java, '-jar', picard_jar, 'CreateSequenceDictionary', 'R=', genome_fna, 'O=', genome_fna[:-4] + '.dict'])
#have_index_files = [_os.path.exists(genome_fna + '.' + a) for a in ('ann','pac','amb','bwt','sa','fai')]
have_index_files = [_os.path.exists(genome_fna + '.' + a) for a in ('fai',)]
if not all(have_index_files):
print('Writing index files for %s' % genome_fna)
_subprocess.call([samtools_exe, 'faidx', genome_fna])
processes = set()
max_processes = _decide_max_processes( max_cpus )
for BAM in self.paths_to_BAMs_dd_si:
intervals = BAM[:-4] + '.intervals'
if not _os.path.exists(intervals) or force:
cmd = [use_java, '-Xmx%sg' % mem_num_gigs, '-jar', jar,
'-T', 'RealignerTargetCreator',
'-R', genome_fna,
'-I', BAM,
'-o', intervals] #, '--validation_strictness', 'LENIENT']
print(' '.join(map(str, cmd)))
processes.add( _subprocess.Popen(cmd, shell=False) )
if len(processes) >= max_processes:
(pid, exit_status) = _os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
else:
print('Found:')
print(intervals)
print('use "force = True" to overwrite')
# Check if all the child processes were closed
for p in processes:
if p.poll() is None:
p.wait()
paths_to_BAMs_dd_si_ra = []
for BAM in self.paths_to_BAMs_dd_si:
intervals = BAM[:-4] + '.intervals'
bam_out = BAM[:-4] + '_realn.bam'
if not _os.path.exists(bam_out) or force:
cmd = [use_java, '-Xmx4g', '-jar', jar,
'-T', 'IndelRealigner',
'-R', genome_fna,
'-I', BAM,
'-targetIntervals', intervals,
'-o', bam_out,
'--filter_bases_not_stored']
print(' '.join(map(str, cmd)))
processes.add( _subprocess.Popen(cmd, shell=False) )
if len(processes) >= max_processes:
_os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
else:
print('Found:')
print(bam_out)
print('use "force = True" to overwrite')
paths_to_BAMs_dd_si_ra += [bam_out]
for p in processes:
if p.poll() is None:
p.wait()
# the last list of BAMs in ready_BAMs is input for CallgVCFsGATK
# both IndelRealignGATK and recalibBaseScoresGATK put here
self.ready_BAMs = [paths_to_BAMs_dd_si_ra]
def align(self, insert_size = False,
path_to_exe = False,
local_alns_path = ['alignments'],
force = False,
max_cpus = -1):
if not path_to_exe:
path_to_exe = _get_exe_path('bwa')
# write genome sequence to a fasta file
try:
_os.makedirs('genome_sequences')
except OSError:
pass
genome_fna = 'genome_sequences/%s.fna' % self.genome_id
_SeqIO.write(_SeqRecord(_Seq(self.genome_sequence.tostring()), id = self.genome_id),
genome_fna,
'fasta')
# make folder for alignments (BAMs)
local_alns_path = _os.path.sep.join(local_alns_path)
if not _os.path.exists(local_alns_path):
_os.makedirs(local_alns_path)
# make a subdir for this genome
local_alns_path_genome = _os.path.sep.join([
local_alns_path,
self.genome_id])
if not _os.path.exists(local_alns_path_genome):
_os.makedirs(local_alns_path_genome)
max_processes = _decide_max_processes( max_cpus )
e1 = 'Could not find "read_files" attribute. Before aligning to genome, reads must be quality score trimmed. Please run trim() method on this Reads instance.'
assert hasattr(self, 'read_files'), e1
e2 = 'Could not find %s. Either run trim() again or ensure file exists'
for pairname, files in self.read_files.items():
assert _os.path.exists(files[1]), e2 % files[1]
assert _os.path.exists(files[2]), e2 % files[2]
have_index_files = [_os.path.exists(genome_fna + '.' + a) for a in ('ann','pac','amb','bwt','sa')]
if not all(have_index_files):
print('Writing BWA index files for %s' % genome_fna)
_subprocess.call([path_to_exe, 'index', genome_fna])
aligned_read_files = {}
for pairname,files in self.read_files.items():
RGinfo = r"@RG\tID:%s\tSM:%s\tPL:ILLUMINA" % (pairname,pairname)
if insert_size:
cmd = [path_to_exe, 'mem', '-t', str(max_processes), '-M', '-a', '-I', insert_size, '-R', RGinfo, genome_fna, files[1], files[2]]
else:
# BWA can estimate on-the-fly
cmd = [path_to_exe, 'mem', '-t', str(max_processes), '-M', '-a', '-R', RGinfo, genome_fna, files[1], files[2]]
out_sam = _os.path.sep.join([local_alns_path_genome, '%s__%s.sam' % (pairname, self.genome_id)])
if not _os.path.exists(out_sam) or force:
print('Called: "%s"' % ' '.join(cmd))
with open(out_sam, "wb") as out:
_subprocess.call(cmd, stdout = out)
else:
print('Found:')
print(out_sam)
print('use "force = True" to overwrite')
print(' '.join(cmd))
aligned_read_files[pairname] = out_sam
self.aligned_read_files = aligned_read_files
def IndelRealignGATK(self,
jar=[
'external_programs', 'GenomeAnalysisTK',
'GenomeAnalysisTK.jar'
],
picard_jar=False,
samtools_exe=False,
use_java='java',
force=False,
mem_num_gigs=2,
max_cpus=-1):
# GATK is manually downloaded by user and placed in folder of their choice
jar = _os.path.sep.join(jar)
if not picard_jar:
picard_jar = _get_jar_path('picard')
if not samtools_exe:
samtools_exe = _get_exe_path('samtools')
genome_fna = 'genome_sequences/{}.fna'.format(self.genome_id)
e1 = 'Could not find "paths_to_BAMs_dd_si" attribute. Before starting GATK analysis, read alignments must have duplicates removed. Please run: .toBAMS(), .removeDuplicates(), .sortIndexBAMs() methods on this SAMs instance, or --deduplicate if using baga_cli.py.'
assert hasattr(self, 'paths_to_BAMs_dd_si'), e1
e2 = 'Could not find %s. Please ensure file exists'
for BAM in self.paths_to_BAMs_dd_si:
assert _os.path.exists(BAM), e2 % BAM
# always (re)generate dict in case of upstream changes in data
print('Creating sequence dictionary for %s' % genome_fna)
_subprocess.call([
use_java, '-jar', picard_jar, 'CreateSequenceDictionary', 'R=',
genome_fna, 'O=', genome_fna[:-4] + '.dict'
])
# always (re)index in case of upstream changes in data
print('Writing index files for %s' % genome_fna)
_subprocess.call([samtools_exe, 'faidx', genome_fna])
processes = set()
max_processes = _decide_max_processes(max_cpus)
for BAM in self.paths_to_BAMs_dd_si:
intervals = BAM[:-4] + '.intervals'
if not _os.path.exists(intervals) or force:
cmd = [
use_java,
'-Xmx%sg' % mem_num_gigs, '-jar', jar, '-T',
'RealignerTargetCreator', '-R', genome_fna, '-I', BAM,
'-o', intervals
] #, '--validation_strictness', 'LENIENT']
print(' '.join(map(str, cmd)))
processes.add(_subprocess.Popen(cmd, shell=False))
if len(processes) >= max_processes:
(pid, exit_status) = _os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
else:
print('Found:')
print(intervals)
print('use "force = True" to overwrite')
# Check if all the child processes were closed
for p in processes:
if p.poll() is None:
p.wait()
paths_to_BAMs_dd_si_ra = []
for BAM in self.paths_to_BAMs_dd_si:
intervals = BAM[:-4] + '.intervals'
bam_out = BAM[:-4] + '_realn.bam'
if not _os.path.exists(bam_out) or force:
cmd = [
use_java, '-Xmx4g', '-jar', jar, '-T', 'IndelRealigner',
'-R', genome_fna, '-I', BAM, '-targetIntervals', intervals,
'-o', bam_out, '--filter_bases_not_stored'
]
print(' '.join(map(str, cmd)))
processes.add(_subprocess.Popen(cmd, shell=False))
if len(processes) >= max_processes:
_os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
else:
print('Found:')
print(bam_out)
print('use "force = True" to overwrite')
paths_to_BAMs_dd_si_ra += [bam_out]
for p in processes:
if p.poll() is None:
p.wait()
# the last list of BAMs in ready_BAMs is input for CallgVCFsGATK
# both IndelRealignGATK and recalibBaseScoresGATK put here
self.ready_BAMs = [paths_to_BAMs_dd_si_ra]
def align(self,
insert_size=False,
path_to_exe=False,
local_alns_path=['alignments'],
force=False,
max_cpus=-1):
if not path_to_exe:
path_to_exe = _get_exe_path('bwa')
# write genome sequence to a fasta file
try:
_os.makedirs('genome_sequences')
except OSError:
pass
genome_fna = 'genome_sequences/%s.fna' % self.genome_id
_SeqIO.write(
_SeqRecord(_Seq(self.genome_sequence.tostring()),
id=self.genome_id), genome_fna, 'fasta')
# make folder for alignments (BAMs)
local_alns_path = _os.path.sep.join(local_alns_path)
if not _os.path.exists(local_alns_path):
_os.makedirs(local_alns_path)
# make a subdir for this genome
local_alns_path_genome = _os.path.sep.join(
[local_alns_path, self.genome_id])
if not _os.path.exists(local_alns_path_genome):
_os.makedirs(local_alns_path_genome)
max_processes = _decide_max_processes(max_cpus)
e1 = 'Could not find "read_files" attribute. Before aligning to genome, reads must be quality score trimmed. Please run trim() method on this Reads instance.'
assert hasattr(self, 'read_files'), e1
e2 = 'Could not find %s. Either run trim() again or ensure file exists'
for pairname, files in self.read_files.items():
assert _os.path.exists(files[1]), e2 % files[1]
assert _os.path.exists(files[2]), e2 % files[2]
# always (re)index in case of upstream changes in data
print('Writing BWA index files for %s' % genome_fna)
_subprocess.call([path_to_exe, 'index', genome_fna])
aligned_read_files = {}
for pairname, files in self.read_files.items():
RGinfo = r"@RG\tID:%s\tSM:%s\tPL:ILLUMINA" % (pairname, pairname)
if insert_size:
cmd = [
path_to_exe, 'mem', '-t',
str(max_processes), '-M', '-a', '-I', insert_size, '-R',
RGinfo, genome_fna, files[1], files[2]
]
else:
# BWA can estimate on-the-fly
cmd = [
path_to_exe, 'mem', '-t',
str(max_processes), '-M', '-a', '-R', RGinfo, genome_fna,
files[1], files[2]
]
out_sam = _os.path.sep.join([
local_alns_path_genome,
'%s__%s.sam' % (pairname, self.genome_id)
])
if not _os.path.exists(out_sam) or force:
print('Called: "%s"' % ' '.join(cmd))
with open(out_sam, "wb") as out:
_subprocess.call(cmd, stdout=out)
else:
print('Found:')
print(out_sam)
print('use "force = True" to overwrite')
print(' '.join(cmd))
aligned_read_files[pairname] = out_sam
self.aligned_read_files = aligned_read_files
def generateReads(self, path_to_exe = False,
paths_to_genomes = False,
readcov = 60,
readlen = 100,
fraglen = 350,
sterrfraglen = 20,
model = 4,
max_cpus = -1):
'''
Call GemSIM to generate reads
Need to have written genome sequences to generate from, possibly with
generated SNPs, small indels and large deletions.
'''
#max_cpus etc
if paths_to_genomes:
use_genomes = sorted(paths_to_genomes)
elif hasattr(self, 'written_genomes'):
use_genomes = sorted(self.written_genomes)
else:
raise ValueError('provide either paths_to_genomes or generate some then .writeSequences()')
if not path_to_exe:
path_to_exe = _get_exe_path('gemsim')
comment2 = '''
to generate reads put GemSIM v1.6 into subfolder GemSIM_v1.6 and issue these commands:
GemSIM_v1.6/GemReads.py -r LESB58_for_GemSim_01.fasta -n 1980527 -l d -u 350 -s 20 -m GemSIM_v1.6/models/ill100v4_p.gzip -c -q 33 -p -o GemSimLESB58_01
'''
num_pairs = len(self.genome.sequence) * readcov / (readlen*2)
if model == 4:
path_to_model = _os.path.sep.join(path_to_exe.split(_os.path.sep)[:-1] + ['models','ill100v4_p.gzip'])
elif model == 5:
path_to_model = _os.path.sep.join(path_to_exe.split(_os.path.sep)[:-1] + ['models','ill100v5_p.gzip'])
print('Using error model: {}'.format(path_to_model))
print('Generating {:,} {}bp read pairs for {}x coverage depth of a {}bp genome ({})'.format(
num_pairs, readlen, readcov, len(self.genome.sequence), self.genome.id))
processes = set()
max_processes = _decide_max_processes( max_cpus )
import time
start = time.time()
out_raw = []
for i,genome_in in enumerate(use_genomes):
# could use per genome length . . less consistent than using reference
# genome_len = len(_SeqIO.read(genome_in,'fasta').seq)
# num_pairs = genome_len * readcov / (readlen*2)
outprefix = 'GemSim_{}_{:02d}'.format(self.genome.id, i+1)
cmd = [path_to_exe,
'-r', genome_in,
'-n', num_pairs,
'-l', 'd', '-u', fraglen, '-s', sterrfraglen,
'-m', path_to_model,
'-c',
'-q', 33, '-p',
'-o', outprefix]
out_raw += [outprefix+'_fir.fastq', outprefix+'_sec.fastq']
# this would be better to rename and compress all in one
# maybe as a shell script? Then resuming (--force) would be easier.
if _os.path.exists(outprefix+'_fir.fastq') and \
_os.path.exists(outprefix+'_sec.fastq'):
print('Found output for {}_fir.fastq (and sec), not regenerating, '\
'delete these to start from scratch'.format(outprefix))
else:
cmd = map(str,cmd)
print(' '.join(cmd))
processes.add( _subprocess.Popen(cmd, shell=False) )
if len(processes) >= max_processes:
(pid, exit_status) = _os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
# Check if all the child processes were closed
for p in processes:
if p.poll() is None:
p.wait()
missing = []
for o in out_raw:
if not _os.path.exists(o):
missing += [o]
assert len(missing) == 0, 'Could not find:\n{}'.format('\n'.join(missing))
print('all finished after {} minutes'.format(int(round((time.time() - start)/60.0))))
outdir = _os.path.sep.join(['simulated_reads',self.genome.id])
try:
_os.makedirs(outdir)
except OSError:
pass
for o in out_raw:
new = _os.path.sep.join([outdir, o.replace('fir','R1').replace('sec','R2')])
print('{} ==> {}'.format(o, new))
_os.rename(o, new)
cmd = ['gzip', new]
print(' '.join(cmd))
_subprocess.call(cmd)
def prep_python_install(extras):
print('Installing via setup.py . . .')
# could try here?
_subprocess.call([_sys.executable, 'setup.py', 'install'] + extras)
def generateReads(self,
path_to_exe=False,
paths_to_genomes=False,
readcov=60,
readlen=100,
fraglen=350,
sterrfraglen=20,
model=4,
max_cpus=-1):
'''
Call GemSIM to generate reads
Need to have written genome sequences to generate from, possibly with
generated SNPs, small indels and large deletions.
'''
#max_cpus etc
if paths_to_genomes:
use_genomes = sorted(paths_to_genomes)
elif hasattr(self, 'written_genomes'):
use_genomes = sorted(self.written_genomes)
else:
raise ValueError(
'provide either paths_to_genomes or generate some then .writeSequences()'
)
if not path_to_exe:
path_to_exe = _get_exe_path('gemsim')
comment2 = '''
to generate reads put GemSIM v1.6 into subfolder GemSIM_v1.6 and issue these commands:
GemSIM_v1.6/GemReads.py -r LESB58_for_GemSim_01.fasta -n 1980527 -l d -u 350 -s 20 -m GemSIM_v1.6/models/ill100v4_p.gzip -c -q 33 -p -o GemSimLESB58_01
'''
num_pairs = len(self.genome.sequence) * readcov / (readlen * 2)
if model == 4:
path_to_model = _os.path.sep.join(
path_to_exe.split(_os.path.sep)[:-1] +
['models', 'ill100v4_p.gzip'])
elif model == 5:
path_to_model = _os.path.sep.join(
path_to_exe.split(_os.path.sep)[:-1] +
['models', 'ill100v5_p.gzip'])
print('Using error model: {}'.format(path_to_model))
print(
'Generating {:,} {}bp read pairs for {}x coverage depth of a {}bp genome ({})'
.format(num_pairs, readlen, readcov, len(self.genome.sequence),
self.genome.id))
processes = set()
max_processes = _decide_max_processes(max_cpus)
import time
start = time.time()
out_raw = []
for i, genome_in in enumerate(use_genomes):
# could use per genome length . . less consistent than using reference
# genome_len = len(_SeqIO.read(genome_in,'fasta').seq)
# num_pairs = genome_len * readcov / (readlen*2)
outprefix = 'GemSim_{}_{:02d}'.format(self.genome.id, i + 1)
cmd = [
path_to_exe, '-r', genome_in, '-n', num_pairs, '-l', 'd', '-u',
fraglen, '-s', sterrfraglen, '-m', path_to_model, '-c', '-q',
33, '-p', '-o', outprefix
]
out_raw += [outprefix + '_fir.fastq', outprefix + '_sec.fastq']
# this would be better to rename and compress all in one
# maybe as a shell script? Then resuming (--force) would be easier.
if _os.path.exists(outprefix+'_fir.fastq') and \
_os.path.exists(outprefix+'_sec.fastq'):
print('Found output for {}_fir.fastq (and sec), not regenerating, '\
'delete these to start from scratch'.format(outprefix))
else:
cmd = map(str, cmd)
print(' '.join(cmd))
processes.add(_subprocess.Popen(cmd, shell=False))
if len(processes) >= max_processes:
(pid, exit_status) = _os.wait()
processes.difference_update(
[p for p in processes if p.poll() is not None])
# Check if all the child processes were closed
for p in processes:
if p.poll() is None:
p.wait()
missing = []
for o in out_raw:
if not _os.path.exists(o):
missing += [o]
assert len(missing) == 0, 'Could not find:\n{}'.format(
'\n'.join(missing))
print('all finished after {} minutes'.format(
int(round((time.time() - start) / 60.0))))
outdir = _os.path.sep.join(['simulated_reads', self.genome.id])
try:
_os.makedirs(outdir)
except OSError:
pass
for o in out_raw:
new = _os.path.sep.join(
[outdir, o.replace('fir', 'R1').replace('sec', 'R2')])
print('{} ==> {}'.format(o, new))
_os.rename(o, new)
cmd = ['gzip', new]
print(' '.join(cmd))
_subprocess.call(cmd)