def make_genome_mapping_index(self):
make_index = False
if ribo_utils.file_exists(
self.settings.get_property('star_genome_dir')):
self.settings.write_to_log(
'STAR index exists at %s' %
self.settings.get_property('star_genome_dir'))
if self.settings.get_property('rebuild_star_index'):
self.settings.write_to_log('STAR index rebuild forced')
make_index = True
else:
self.settings.write_to_log('using existing STAR index')
else:
make_index = True
ribo_utils.make_dir(self.settings.get_property('star_genome_dir'))
if make_index:
self.settings.write_to_log('building STAR index')
if self.settings.get_property('transcriptome_mapping_only'):
command_to_run = 'STAR --runThreadN %d --runMode genomeGenerate --genomeDir %s --genomeFastaFiles %s --genomeSAsparseD %d 1>>%s 2>>%s' % (
self.threads,
self.settings.get_property('star_genome_dir'),
self.settings.get_transcriptome_FASTA(),
self.settings.get_property('star_index_sparsity'),
self.settings.get_log(), self.settings.get_log())
else:
command_to_run = 'STAR --runThreadN %d --runMode genomeGenerate --genomeDir %s --genomeFastaFiles %s*.fa --genomeSAsparseD %d 1>>%s 2>>%s' % (
self.threads,
self.settings.get_property('star_genome_dir'),
self.settings.get_genome_sequence_dir(),
self.settings.get_property('star_index_sparsity'),
self.settings.get_log(), self.settings.get_log())
self.settings.write_to_log(command_to_run)
subprocess.Popen(command_to_run, shell=True).wait()
self.settings.write_to_log('STAR index ready')
def sequence_counts_exist(self):
sequence_counts = self.get_transcript_counts()
return ribo_utils.file_exists(sequence_counts)
def process_settings(self, settings_file):
"""
- reads the settings file and converts str to float, list, etc.
- stores result in self.settings as a dict()
- CRITICAL NOTE: All keys must be lower case
"""
# TODO: Add new parameters and comments to settings files
int_keys = [
'comparison_read_cutoff', 'min_post_trimming_length',
'max_post_trimming_length', 'sequence_quality_cutoff', 'trim_5p',
'star_index_sparsity', 'outfiltermultimapnmax',
'outfiltermismatchnmax', 'alignsjdboverhangmin',
'alignsjoverhangmin', 'alignintronmax', 'five_prime_p_offset'
]
float_keys = ['atail_purity_cutoff']
str_keys = [
'adaptor_3p_sequence', 'star_genome_dir', 'star_ncrna_dir',
'genome_sequence_dir', 'ncrna_sequence_dir', 'annotation_gtf_file',
'qc_annotation_gtf_file', 'alignendstype'
]
#if transcriptome_only is true, will genreate a transcriptome-only genome to map to
boolean_keys = [
'transcriptome_mapping_only', 'deduplicate_reads',
'force_remapping', 'force_recount', 'rebuild_star_index',
'force_retrim', 'make_interactive_plots', 'reads_reversed',
'add_3_for_stop', 'forbid_non_polya_soft_clip',
'unique_mapping_only'
]
list_str_keys = ['fastq_gz_files', 'sample_names']
#list_float_keys = ['concentrations', 'input_rna']
extant_files = ['genome_sequence_dir', 'annotation_gtf_file']
config = ConfigParser.ConfigParser()
config.read(settings_file)
settings = {}
for section in config.sections():
for option in config.options(section):
settings[option] = config.get(section, option)
settings[section] = True
for k in int_keys:
settings[k] = int(settings[k])
for k in str_keys:
settings[k] = settings[k]
for k in float_keys:
settings[k] = float(settings[k])
for k in boolean_keys:
if not settings[k].lower() in ['true', 'false']:
raise ValueError('Boolean value %s must be "true" or "false"' %
k)
settings[k] = settings[k].lower() == 'true'
#for k in list_float_keys:
# settings[k] = map(float, simplejson.loads(settings[k]))
#for k in list_int_keys:
# settings[k] = map(int, simplejson.loads(settings[k]))
for k in list_str_keys:
settings[k] = simplejson.loads(settings[k])
self.fqdir = settings['fastq_dir']
self.sample_names = settings['sample_names']
self.fastq_gz_files = settings['fastq_gz_files']
self.fastq_gz_file_handles = [
os.path.join(self.fqdir, fastq_gz_file)
for fastq_gz_file in self.fastq_gz_files
]
for file_handle in self.fastq_gz_file_handles:
try:
assert ribo_utils.file_exists(file_handle)
except:
print 'ERROR: nonexistent file ', file_handle
sys.exit()
for k in extant_files:
try:
assert ribo_utils.file_exists(settings[k])
except AssertionError:
print 'file %s does not exist' % settings[k]
sys.exit()
self.settings = settings
self.rdir = settings['results_dir']
ribo_utils.make_dir(self.rdir)
shutil.copy(settings_file, self.rdir)
def genome_mapped_reads_exist(self):
mapped_reads = self.get_genome_mapped_reads()
return ribo_utils.file_exists(mapped_reads)
def qc_pickle_exists(self):
qc_pickle = self.get_qc_pickle()
return ribo_utils.file_exists(qc_pickle)
def ncrna_unmapped_reads_exist(self):
mapped_reads = self.get_ncrna_unmapped_reads()
return ribo_utils.file_exists(mapped_reads)
def ncrna_mapping_finished(self):
unmapped_reads_summary = os.path.join(
self.experiment_settings.get_rdir(), 'ncrna_mapped_reads',
'%(sample_name)sLog.final.out' % {'sample_name': self.sample_name})
return ribo_utils.file_exists(unmapped_reads_summary)
def trimmed_reads_exist(self):
trimmed_reads = self.get_trimmed_reads()
return ribo_utils.file_exists(trimmed_reads)
def deduplicated_reads_exist(self):
deduplicated_reads = self.get_deduplicated_reads()
return ribo_utils.file_exists(deduplicated_reads)
def adaptorless_reads_exist(self):
trimmed_reads = self.get_adaptor_trimmed_reads()
return ribo_utils.file_exists(trimmed_reads)
