def rsem_index(rsem_index_executable, fasta_input, bowtie_info, params):
'''
This module will create the rsem indexes at params.index_destination using
RSEM_INDEX_EXECUTABLE. If FASTA_INPUT = True, it will use the bowtie version
to make bowtie indexes as well.
bowtie_info is a tuple of (bowtie_path, bowtie_version)
params contains
index_destination - Folder to store the indexes
n - number of cores to use
genome_fasta - path to genomic fasta file. Can also specify DOWNLOAD.
genome_version - hg19/hg38
logfile - Open file handle to a log file
RETURN VALUES
index_path - Path to directory where nidexes were stored
'''
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Creating rsem references...',
file=params.logfile)
index_path = os.path.abspath(params.index_destination)
# If the directory doesn't exist, create it
if not os.path.exists(index_path):
prepare.py_mkdir(index_path)
if params.genome_fasta == 'DOWNLOAD':
params.genome_fasta = prepare.get_genome(params.genome_version,
index_path,
params.tbtf_executable,
params.logfile)
else:
params.genome_fasta = pi_errors.test_param_value(
params.genome_fasta, 'Genomic Fasta', '--genome_fasta',
params.logfile)
# If the gtf file is required, download it
gencode_file = prepare.get_gtf(params.genome_version, index_path,
params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Running rsem-prepare-reference on fasta reference.',
file=params.logfile)
rsem_prepref_call = [rsem_index_executable] # base call
rsem_prepref_call.extend(['--gtf', gencode_file]) # gtf file
if fasta_input:
rsem_prepref_call.extend([
''.join(['--', bowtie_version]),
''.join(['--', bowtie_version, '-path']), bowtie_path
])
else:
rsem_prepref_call.append('--no-bowtie')
rsem_prepref_call.append(params.genome_fasta)
rsem_prepref_call.extend(
[''.join([index_path, '/', params.genome_version])])
print(rsem_prepref_call, file=params.logfile)
return_value = call(rsem_prepref_call)
if return_value != 0:
raise pi_errors.MyRuntimeError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Indexing Failed', params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing completed.',
file=params.logfile)
return index_path
def rsem_index(rsem_index_executable, fasta_input, bowtie_info, params):
'''
This module will create the rsem indexes at params.index_destination using
RSEM_INDEX_EXECUTABLE. If FASTA_INPUT = True, it will use the bowtie version
to make bowtie indexes as well.
bowtie_info is a tuple of (bowtie_path, bowtie_version)
params contains
index_destination - Folder to store the indexes
n - number of cores to use
genome_fasta - path to genomic fasta file. Can also specify DOWNLOAD.
genome_version - hg19/hg38
logfile - Open file handle to a log file
RETURN VALUES
index_path - Path to directory where nidexes were stored
'''
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Creating rsem references...', file=params.logfile)
index_path = os.path.abspath(params.index_destination)
# If the directory doesn't exist, create it
if not os.path.exists(index_path):
prepare.py_mkdir(index_path)
if params.genome_fasta == 'DOWNLOAD':
params.genome_fasta = prepare.get_genome(params.genome_version,
index_path,
params.tbtf_executable,
params.logfile)
else:
params.genome_fasta = pi_errors.test_param_value(params.genome_fasta,
'Genomic Fasta',
'--genome_fasta',
params.logfile)
# If the gtf file is required, download it
gencode_file = prepare.get_gtf(params.genome_version, index_path,
params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Running rsem-prepare-reference on fasta reference.',
file=params.logfile)
rsem_prepref_call = [rsem_index_executable] # base call
rsem_prepref_call.extend(['--gtf', gencode_file]) # gtf file
if fasta_input:
rsem_prepref_call.extend([''.join(['--', bowtie_version]),
''.join(['--', bowtie_version, '-path']),
bowtie_path])
else:
rsem_prepref_call.append('--no-bowtie')
rsem_prepref_call.append(params.genome_fasta)
rsem_prepref_call.extend([''.join([index_path, '/',
params.genome_version])])
print(rsem_prepref_call, file=params.logfile)
return_value = call(rsem_prepref_call)
if return_value != 0:
raise pi_errors.MyRuntimeError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Indexing Failed', params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing completed.', file=params.logfile)
return index_path
def star_indexing(star_executable, read_length, params):
'''
This module indexes a genome using STAR_EXECUTABLE using READ_LENGTH to set
edge size.
params contains
index_destination - The location where the index should be stored
logfile - Open file handle to a log file
genome_version - hg19/hg38
n - number of cores to use
tbtf_executable - path to twoBitToFa
RETURN VALUES
index_path - path ot the directory where indexes were stored
'''
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing fasta...',
file=params.logfile)
params.index_destination = os.path.abspath(params.index_destination)
if not os.path.exists(params.index_destination):
prepare.py_mkdir(params.index_destination)
edge_size = max(50, int(round(read_length / 50, 0) * 50)) # minimum edge
# size = 50
index_path = ''.join(
[params.index_destination, '/STAR_',
str(edge_size), '_references'])
if not os.path.exists(index_path): # make reference based on edge size
prepare.py_mkdir(index_path)
genome_fasta = prepare.get_genome(params.genome_version, index_path,
params.tbtf_executable, params.logfile)
gencode_file = prepare.get_gtf(params.genome_version, index_path,
params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Running STAR index on fasta reference.',
file=params.logfile)
starindex_call = [star_executable] # Base call
starindex_call.extend(['--runThreadN', str(params.n)]) # Threads
starindex_call.extend(['--runMode', 'genomeGenerate']) # Indexing module
starindex_call.extend(['--genomeDir', index_path]) # index directory
starindex_call.extend(['--genomeFastaFiles', genome_fasta]) # Genomic fa
starindex_call.extend(['--sjdbGTFfile', gencode_file]) # gencode annots
starindex_call.extend(['--sjdbOverhang', str(read_length)]) # edge size
print(starindex_call, file=params.logfile)
return_value = call(starindex_call)
if return_value != 0:
raise pi_errors.MyRuntimeError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Indexing Failed', params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing completed.',
file=params.logfile)
return index_path
def star_indexing(star_executable, read_length, params):
'''
This module indexes a genome using STAR_EXECUTABLE using READ_LENGTH to set
edge size.
params contains
index_destination - The location where the index should be stored
logfile - Open file handle to a log file
genome_version - hg19/hg38
n - number of cores to use
tbtf_executable - path to twoBitToFa
RETURN VALUES
index_path - path ot the directory where indexes were stored
'''
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing fasta...', file=params.logfile)
params.index_destination = os.path.abspath(params.index_destination)
if not os.path.exists(params.index_destination):
prepare.py_mkdir(params.index_destination)
edge_size = max(50, int(round(read_length / 50, 0) * 50)) # minimum edge
# size = 50
index_path = ''.join([params.index_destination, '/STAR_',
str(edge_size), '_references'])
if not os.path.exists(index_path): # make reference based on edge size
prepare.py_mkdir(index_path)
genome_fasta = prepare.get_genome(params.genome_version, index_path,
params.tbtf_executable,
params.logfile)
gencode_file = prepare.get_gtf(params.genome_version, index_path,
params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Running STAR index on fasta reference.', file=params.logfile)
starindex_call = [star_executable] # Base call
starindex_call.extend(['--runThreadN', str(params.n)]) # Threads
starindex_call.extend(['--runMode', 'genomeGenerate']) # Indexing module
starindex_call.extend(['--genomeDir', index_path]) # index directory
starindex_call.extend(['--genomeFastaFiles', genome_fasta]) # Genomic fa
starindex_call.extend(['--sjdbGTFfile', gencode_file]) # gencode annots
starindex_call.extend(['--sjdbOverhang', str(read_length)]) # edge size
print(starindex_call, file=params.logfile)
return_value = call(starindex_call)
if return_value != 0:
raise pi_errors.MyRuntimeError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Indexing Failed', params.logfile)
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') + ': ' +
'Indexing completed.', file=params.logfile)
return index_path
def process_parameters(params):
'''
This module conducts the error handling for all parmeters passed to the
program.
'''
print('PROGRESS ' + dt.now().strftime('%I:%M %p %b %d, %Y') +
': Processing input parameters.', file=params.logfile)
# Does the input vcf file exist?
if not os.path.exists(''.join([params.file_path, '/', params.file_prefix,
'.vcf'])):
raise pi_errors.InputFileError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Please provide a valid input file using --file_prefix',
params.logfile)
# The memory option for java should be of the form Xmx10G or Xmx10M
if not (params.java_Xmx.endswith('G') or params.java_Xmx.endswith('M')):
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': Please use a suitable value for --Xmx.', params.logfile)
params.java_executable = pi_errors.test_param_value(params.java_executable,
'java',
'--java',
params.logfile)
# Does the provided snpeff binary provided exist?
params.snpeff_jar = pi_errors.test_param_value(params.snpeff_jar,
'snpeff',
'--snpeff_jar',
params.logfile)
params.use_snpeff_db = False
# Does the user want a snpEff packaged database?
if params.config_file == 'PACKAGED':
params.use_snpeff_db = True
# Has the snpeff reference to be used been provided?
if params.reference_name == None:
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': --snp_reference is required if --config=PACKAGED.',
params.logfile)
# If a custom databse is desired, does it need to be created?
if params.index_location is None:
# If the user has provided the location to the parent directory of data
# directory, make DATA_DIRECTORY point to data. If they have provided
# the link to data, make INDEX_DESTINATION point to the parent and
# DATA_DIRECTORY point to data.
if os.path.split(params.index_destination.rstrip('/'))[1] != 'data':
params.data_directory = '/'.join([params.index_destination, 'data'])
else:
params.data_directory = params.index_destination
params.index_destination = \
params.index_destination.rstrip('/').rstrip('/data')
# Create the data directory if needed
if not os.path.exists(params.data_directory):
prepare.py_mkdir(params.data_directory)
# If we're using a custom databse, thre is nothing more to do
if params.use_snpeff_db:
return None
# Initialise the reference name
params.reference_name = ''.join([params.genome_version, '_custom'])
# make a variable to gold GENOME_VERSION_custom
genome_folder = '/'.join([params.data_directory, params.reference_name])
prepare.py_mkdir(genome_folder)
# If the genome fasta isn't provided or is provided a wrong value,
# download it
if params.genome_fasta == 'DOWNLOAD' or not \
os.path.exists(params.genome_fasta):
# Does the provided tbtf binary point to a valid file?
params.tbtf_executable = pi_errors.test_param_value(
params.tbtf_executable, 'twoBitToFa', '--twoBitToFa',
params.logfile)
params.genome_fasta = prepare.get_genome(
params.genome_version, genome_folder,
params.tbtf_executable, params.logfile)
# Rename genome fasta
call(['mv', params.genome_fasta, '/'.join([genome_folder,
'sequences.fa'])])
else:
params.genome_fasta = os.path.abspath(params.genome_fasta)
# Link sequencesfa to genome fasta
call(['ln', '-s', '-T', params.genome_fasta,
'/'.join([genome_folder, 'sequences.fa'])])
# Download the gencode GTF file
params.gtf_file = prepare.get_gtf(params.genome_version,
genome_folder,
params.logfile)
# Rename gtf file
call(['mv', params.gtf_file, '/'.join([genome_folder, 'genes.gtf'])])
# If it has been provided, set up the config file
else:
# If the user has provided the location to the parent directory of data
# directory, make DATA_DIRECTORY point to data. If they have provided
# the link to data, make INDEX_LOCATION point to the parent and
# DATA_DIRECTORY point to data.
if os.path.split(params.index_location.rstrip('/'))[1] != 'data':
params.data_directory = '/'.join([params.index_location, 'data'])
else:
params.data_directory = params.index_location
params.index_location = \
params.index_location.rstrip('/').rstrip('/data')
# If we're using a custom databse, thre is nothing more to do
if params.use_snpeff_db:
return None
# If the config file hasn't been provided, is it in INDEX_LOCATION
# AND does GENOME_VERSION_custom exist (i.e. was it created by
# this script?)
if params.config_file is None:
params.config_file = pi_errors.test_param_value(
'/'.join([params.index_location, 'snpEff.config']),
'snpEff.config', '--config', params.logfile)
# Dummy variable to ensure the GENOME_VERSION_custom exists
_ = pi_errors.test_param_value(
''.join([params.data_directory, '/', params.genome_version,
'_custom']), '_'.join([params.genome_version, 'custom'
]), '--snpeff_reference and' +
'--config', params.logfile)
params.reference_name = '_'.join([params.genome_version, 'custom'])
# If a config file has been provided, does it point to a legit file and
# has the reference name also been provided?
else:
params.config_file = pi_errors.test_param_value(
params.config_file, 'snpEff config file', '--config',
params.logfile)
if params.reference_name is None:
raise pi_errors.ParameterError(
dt.now().strftime('%I:%M %p %b %d, %Y') + \
': --snpeff_reference is required if --config points to' + \
' a custom file.', params.logfile)
return None