def run(project):
logging.info('Beginning featureCounts component of pipeline.')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
config_parser = component_utils.load_remote_module('config_parser', utils_dir)
util_methods = component_utils.load_remote_module('util_methods', utils_dir)
# parse this module's config file
component_utils.parse_config_file(project, util_methods, config_parser, os.path.dirname(os.path.realpath(__file__)))
# create a full path to the output directory for the featureCount's output and reset the parameter in the project parameters:
output_dir = os.path.join(project.parameters.get('output_location'), project.parameters.get('feature_counts_output_dir'))
project.parameters.reset_param('feature_counts_output_dir', output_dir)
# create the final output directory, if possible
util_methods.create_directory(output_dir)
# start the counting:
execute_counting(project, util_methods)
# create the final, unnormalized count matrices for each set of BAM files
merged_count_files = create_count_matrices(project, util_methods)
# add to the project object:
project.count_matrices = merged_count_files
def run(project):
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
config_parser = component_utils.load_remote_module('config_parser',
utils_dir)
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
component_utils.parse_config_file(
project, util_methods, config_parser,
os.path.dirname(os.path.realpath(__file__)))
# create a full path to the output directory for the output and reset the parameter in the project parameters:
output_dir = os.path.join(
project.parameters.get('output_location'),
project.parameters.get('normalized_counts_output_dir'))
project.parameters.reset_param('normalized_counts_output_dir', output_dir)
# create the final output directory, if possible
util_methods.create_directory(output_dir)
# perform the actual normalization:
normalize(project)
def run(project):
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
config_parser = component_utils.load_remote_module('config_parser',
utils_dir)
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
def run(name, project):
logging.info('Beginning DESeq differential expression analysis...')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the util_methods module:
util_methods = component_utils.load_remote_module('util_methods', utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(project, this_dir))
component_params = component_utils.parse_config_file(project, this_dir, 'COMPONENT_SPECIFIC')
# create a full path to the output directory for the output:
output_dir = os.path.join(project.parameters.get('output_location'), component_params.get('deseq_output_dir'))
component_params['deseq_output_dir'] = output_dir
# create the final output directory, if possible
util_methods.create_directory(output_dir, overwrite = True)
deseq_output_files, heatmap_files = call_deseq(project, component_params)
# write a summary of the number of differentially expressed genes
create_diff_exp_summary(deseq_output_files, project, component_params)
# change permissions:
[os.chmod(f,0775) for f in deseq_output_files.values()]
def run(name, project):
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the util_methods module:
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(
project, this_dir))
component_params = component_utils.parse_config_file(
project, this_dir, 'COMPONENT_SPECIFIC')
# create a full path to the output directory for the output and reset the parameter in the project parameters:
output_dir = os.path.join(
project.parameters.get('output_location'),
component_params.get('normalized_counts_output_dir'))
component_params['normalized_counts_output_dir'] = output_dir
# create the final output directory, if possible
util_methods.create_directory(output_dir, overwrite=True)
# perform the actual normalization:
output_files = normalize(project, component_params)
logging.info('Done with normalize. Output files: %s' % output_files)
# change permissions on those output files:
[os.chmod(f, 0775) for f in output_files.values()]
def run(name, project):
logging.info('Beginning creation of custom latex report')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the util_methods module:
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(
project, this_dir))
component_params = component_utils.parse_config_file(
project, this_dir, 'COMPONENT_SPECIFIC')
# TODO: generate report when we didn't align-- use samtools, etc. to get alignment stats
if project.parameters.get(
'aligner') == 'star' and not project.parameters.get('skip_align'):
extra_params = component_utils.parse_config_file(
project, this_dir, 'STAR')
else:
raise PdfReportNotConfiguredForAlignerException(
'Please configure the pdf report generator for this particular aligner.'
)
# create a full path to the output directory for the output:
output_dir = os.path.join(project.parameters.get('output_location'),
component_params.get('report_output_dir'))
component_params['report_output_dir'] = output_dir
# create the final output directory, if possible
util_methods.create_directory(output_dir, overwrite=True)
# read template
env = jinja2.Environment(loader=jinja2.FileSystemLoader(this_dir))
template = env.get_template(component_params.get('report_template'))
output_file = create_report(template, project, component_params,
extra_params)
# change permissions:
os.chmod(output_file, 0775)
# have to give the constructor of ComponentOutput a dictionary
output_file_map = {os.path.basename(output_file): output_file}
# create the ComponentOutput object and return it
logging.info('Preparing ComponentOutput for latex report module')
c1 = component_utils.ComponentOutput(
output_file_map, component_params.get('report_tab_title'),
component_params.get('report_header_msg'),
component_params.get('report_display_format'))
return [c1]
def run(name, project):
logging.info('Beginning GSEA component.')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(
project, this_dir))
component_params = component_utils.parse_config_file(
project, this_dir, 'COMPONENT_SPECIFIC')
if project.parameters.get('genome') in component_params.get(
'acceptable_genomes'):
# create a full path to the output directory for GSEA's output:
output_dir = os.path.join(project.parameters.get('output_location'),
component_params.get('gsea_output_dir'))
component_params['gsea_output_dir'] = output_dir
# create the final output directory, if possible
logging.info('Creating output directory at %s' % output_dir)
util_methods.create_directory(output_dir, overwrite=True)
# create the cls and gct files for input to GSEA:
logging.info('About to create the CLS and GCT files')
create_input_files(project, component_params)
# run it:
logging.info('Actually run GSEA')
output = run_gsea(project, component_params, util_methods)
return [
component_utils.ComponentOutput(
output, component_params.get('tab_title'),
component_params.get('header_msg'),
component_params.get('display_format')),
]
else:
# return a list of None if we shouldn't run GSEA for this genome
return [
None,
]
def run(name, project):
logging.info('Beginning featureCounts component of pipeline.')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(
project, this_dir))
component_params = component_utils.parse_config_file(
project, this_dir, 'COMPONENT_SPECIFIC')
# create a full path to the output directory for the featureCount's output:
output_dir = os.path.join(
project.parameters.get('output_location'),
component_params.get('feature_counts_output_dir'))
component_params['feature_counts_output_dir'] = output_dir
# create the final output directory, if possible
util_methods.create_directory(output_dir, overwrite=True)
# start the counting:
execute_counting(project, component_params, util_methods)
# create the final, unnormalized count matrices for each set of BAM files
merged_count_files = create_count_matrices(project, component_params,
util_methods)
# add these common files to the project object (so that other components have access to them):
project.raw_count_matrices = merged_count_files
# change permissions on all those:
[os.chmod(f, 0775) for f in merged_count_files]
def run(name, project):
logging.info('Beginning rnaSeQC component of pipeline.')
# get the location of the utils directory:
utils_dir = project.parameters.get('utils_dir')
# load the parser and the util_methods modules:
util_methods = component_utils.load_remote_module('util_methods',
utils_dir)
# parse this module's config file
this_dir = os.path.dirname(os.path.realpath(__file__))
project.parameters.add(component_utils.parse_config_file(
project, this_dir))
component_params = component_utils.parse_config_file(
project, this_dir, 'COMPONENT_SPECIFIC')
component_params.update(
component_utils.parse_config_file(project, this_dir,
project.parameters.get('genome')))
# create a full path to the output directory for rnaseQC's output:
output_dir = os.path.join(project.parameters.get('output_location'),
component_params.get('rnaseqc_output_dir'))
# create the final output directory, if possible
util_methods.create_directory(output_dir, overwrite=True)
# run the QC processes:
reports = run_qc(project, component_params, util_methods)
return [
component_utils.ComponentOutput(
reports, component_params.get('tab_title'),
component_params.get('header_msg'),
component_params.get('display_format')),
]