def __init__(self,
exp_seqs_fname,
control_seqs_fname,
kmer_lens,
output_dir,
exp_coords_fname=None,
control_coords_fname=None,
genome="mm9"):
# FASTAs representing the sequences for exp and control
# conditions
self.exp_seqs_fname = exp_seqs_fname
self.control_seqs_fname = control_seqs_fname
# Coordinates files for exp and control conditions
self.exp_coords_fname = exp_coords_fname
self.control_coords_fname = control_coords_fname
# Kmer lengths to consider
self.kmer_lens = kmer_lens
self.output_dir = output_dir
# Optional genome name
self.genome = genome
# Define Kmer tables for each kmer length
self.exp_kmer_tables = []
self.control_kmer_tables = []
self.logger = utils.get_logger("MotifSet",
os.path.join(self.output_dir, "logs"))
def __init__(self,
exp_seqs_fname,
control_seqs_fname,
kmer_lens,
output_dir,
exp_coords_fname=None,
control_coords_fname=None,
genome="mm9"):
# FASTAs representing the sequences for exp and control
# conditions
self.exp_seqs_fname = exp_seqs_fname
self.control_seqs_fname = control_seqs_fname
# Coordinates files for exp and control conditions
self.exp_coords_fname = exp_coords_fname
self.control_coords_fname = control_coords_fname
# Kmer lengths to consider
self.kmer_lens = kmer_lens
self.output_dir = output_dir
# Optional genome name
self.genome = genome
# Define Kmer tables for each kmer length
self.exp_kmer_tables = []
self.control_kmer_tables = []
self.logger = utils.get_logger("MotifSet",
os.path.join(self.output_dir, "logs"))
def __init__(self, sample, pipeline):
# Pipeline instance that the sample is attached to
self.pipeline = pipeline
self.sample = sample
self.settings_info = pipeline.settings_info
# Define logger
self.logger = utils.get_logger("QualityControl.%s" % (sample.label), self.pipeline.pipeline_outdirs["logs"])
# QC header: order of QC fields to be outputted
self.regions_header = [
"num_ribo",
"num_exons",
"num_cds",
"num_introns",
"num_3p_utr",
"num_5p_utr",
"num_tRNAs",
"num_junctions",
]
self.qc_stats_header = [
"percent_mapped",
"percent_unique",
"percent_ribo",
"percent_exons",
"percent_cds",
"percent_introns",
"percent_3p_utr",
"percent_5p_utr",
"percent_tRNAs",
"3p_to_cds",
"5p_to_cds",
"3p_to_5p",
"exon_intron_ratio",
]
self.qc_header = (
["num_reads", "num_mapped", "num_ribosub_mapped", "num_unique_mapped"]
+ self.qc_stats_header
+ self.regions_header
)
# QC results
self.na_val = "NA"
self.qc_results = defaultdict(lambda: self.na_val)
# QC output dir
self.qc_outdir = self.pipeline.pipeline_outdirs["qc"]
# QC filename for this sample
self.sample_outdir = os.path.join(self.qc_outdir, self.sample.label)
utils.make_dir(self.sample_outdir)
# Regions output dir
self.regions_outdir = os.path.join(self.sample_outdir, "regions")
utils.make_dir(self.regions_outdir)
self.qc_filename = os.path.join(self.sample_outdir, "%s.qc.txt" % (self.sample.label))
self.qc_loaded = False
# use ensGene gene table for QC computations
self.gene_table = self.pipeline.rna_base.gene_tables["ensGene"]
# Load QC information if file corresponding to sample
# already exists
self.load_qc_from_file()
def __init__(self, settings_filename, log_output_dir, curr_sample=None):
"""
Initialize pipeline.
"""
# If invoked to run on particular sample
self.curr_sample = curr_sample
self.genome = None
# Output directory for logging pipeline activity
self.log_output_dir = log_output_dir
# Output directory for actual pipeline output
self.output_dir = None
# Load settings file
self.settings_filename = settings_filename
# Load settings
self.sequence_filenames = None
self.parsed_settings = None
self.settings_info = None
self.data_type = None
# Directory where pipeline init files are
self.init_dir = None
# Paired-end or not
self.is_paired_end = None
self.sample_to_group = None
self.group_to_samples = None
self.samples = []
# Cluster objects to use
self.my_cluster = None
# Check settings are correct
self.load_pipeline_settings()
# Pipeline output subdirectories
self.pipeline_outdirs = {}
# RPKM directory for teh pipeline
self.rpkm_dir = None
# QC objects for each sample in pipeline
self.qc_objects = {}
# Top-level output dirs
self.toplevel_dirs = ["rawdata", "mapping", "qc", "analysis", "logs"]
self.init_outdirs()
pipeline_log_name = "Pipeline"
if self.curr_sample is not None:
pipeline_log_name = "Pipeline.%s" % (self.curr_sample)
self.logger = utils.get_logger(pipeline_log_name, self.pipeline_outdirs["logs"])
self.load_cluster()
## Load RNA Base: object storing all the relevant
## initialization information
self.rna_base = None
self.load_rna_base()
## Load samples
self.load_pipeline_samples()
## Initialize QC for samples
# QC header: order of QC fields to be outputted
self.qc_header = []
self.init_qc()
self.na_val = "NA"
def __init__(self, sample, pipeline):
# Pipeline instance that the sample is attached to
self.pipeline = pipeline
self.sample = sample
self.settings_info = pipeline.settings_info
# Define logger
self.logger = utils.get_logger("QualityControl.%s" % (sample.label),
self.pipeline.pipeline_outdirs["logs"])
# QC header: order of QC fields to be outputted
self.regions_header = [
"num_ribo", "num_exons", "num_cds", "num_introns", "num_3p_utr",
"num_5p_utr", "num_tRNAs", "num_junctions"
]
self.qc_stats_header = [
"percent_mapped", "percent_unique", "percent_ribo",
"percent_exons", "percent_cds", "percent_introns",
"percent_3p_utr", "percent_5p_utr", "percent_tRNAs", "3p_to_cds",
"5p_to_cds", "3p_to_5p", "exon_intron_ratio"
]
self.qc_header = ["num_reads",
"num_mapped",
"num_ribosub_mapped",
"num_unique_mapped"] + \
self.qc_stats_header + \
self.regions_header
# QC results
self.na_val = "NA"
self.qc_results = defaultdict(lambda: self.na_val)
# QC output dir
self.qc_outdir = self.pipeline.pipeline_outdirs["qc"]
# QC filename for this sample
self.sample_outdir = os.path.join(self.qc_outdir, self.sample.label)
utils.make_dir(self.sample_outdir)
# Regions output dir
self.regions_outdir = os.path.join(self.sample_outdir, "regions")
utils.make_dir(self.regions_outdir)
self.qc_filename = os.path.join(self.sample_outdir,
"%s.qc.txt" % (self.sample.label))
self.qc_loaded = False
# use ensGene gene table for QC computations
self.gene_table = self.pipeline.rna_base.gene_tables["ensGene"]
# Load QC information if file corresponding to sample
# already exists
self.load_qc_from_file()
def __init__(self, results_dir, output_dir,
label=None):
"""
Load up results directory
"""
self.output_dir = output_dir
self.logger_label = label
if self.logger_label is None:
self.logger_label = "BindnSeq"
self.logger = utils.get_logger(self.logger_label,
self.output_dir)
self.results_dir = results_dir
self.label = label
# All kmer lengths to load
self.kmer_lens = [4, 5, 6, 7, 8, 9]
# Odds ratios (DataFrames indexed by kmer length)
self.odds_ratios = {}
# Counts (DataFrames indexed by kmer length)
self.counts = {}
def load_settings(self):
"""
Load settings for misowrap.
"""
settings_info, parsed_settings = \
misowrap_settings.load_misowrap_settings(self.settings_filename)
self.settings_info = settings_info
# Load basic settings about data
self.read_len = self.settings_info["settings"]["readlen"]
self.overhang_len = self.settings_info["settings"]["overhanglen"]
self.miso_bin_dir = \
utils.pathify(self.settings_info["settings"]["miso_bin_dir"])
self.miso_settings_filename = \
utils.pathify(self.settings_info["settings"]["miso_settings_filename"])
self.miso_events_dir = \
utils.pathify(self.settings_info["settings"]["miso_events_dir"])
self.miso_outdir = \
utils.pathify(self.settings_info["settings"]["miso_output_dir"])
# Load data-related parameters
self.bam_files = self.settings_info["data"]["bam_files"]
if "insert_lens_dir" in self.settings_info["data"]:
self.insert_lens_dir = \
utils.pathify(self.settings_info["data"]["insert_lens_dir"])
# Sample labels
self.sample_labels = self.settings_info["data"]["sample_labels"]
# Set output directories
self.comparisons_dir = os.path.join(self.output_dir,
"comparisons")
self.comparison_groups = \
self.settings_info["data"]["comparison_groups"]
self.logs_outdir = os.path.join(self.output_dir,
"misowrap_logs")
# Create necessary directories
utils.make_dir(self.miso_outdir)
utils.make_dir(self.comparisons_dir)
utils.make_dir(self.logs_outdir)
if "cluster_type" in self.settings_info["settings"]:
self.use_cluster = True
self.cluster_type = \
self.settings_info["settings"]["cluster_type"]
self.chunk_jobs = \
self.settings_info["settings"]["chunk_jobs"]
if self.use_cluster:
print "Loading cluster information."
# Load cluster object if given a cluster type
self.load_cluster()
# Create a logger object
if self.logger_label is None:
self.logger_label = "misowrap"
else:
self.logger_label = "misowrap_%s" %(logger_label)
self.logger = utils.get_logger(self.logger_label,
self.logs_outdir)
# Whether to prefilter MISO events
# Set general default settings
if "prefilter_miso" not in settings_info["settings"]:
# By default, set it so that MISO events are not
# prefiltered
settings_info["settings"]["prefilter_miso"] = False
self.prefilter_miso = \
self.settings_info["settings"]["prefilter_miso"]
# Load event types
self.load_event_types()
# Set path to MISO scripts
self.run_miso_cmd = os.path.join(self.miso_bin_dir,
"run_miso.py")
self.run_events_cmd = os.path.join(self.miso_bin_dir,
"run_events_analysis.py")
self.pe_utils_cmd = os.path.join(self.miso_bin_dir,
"pe_utils.py")
# Files related to gene tables
self.tables_dir = \
os.path.join(self.settings_info["pipeline-files"]["init_dir"],
"ucsc")
if not os.path.isdir(self.tables_dir):
print "Error: %s directory does not exist." \
%(self.tables_dir)
sys.exit(1)
self.const_exons_gff = os.path.join(self.tables_dir,
"exons",
"const_exons",
"ensGene.const_exons.gff")
if not os.path.isfile(self.const_exons_gff):
print "Error: Const. exons GFF %s does not exist." \
%(self.const_exons_gff)
sys.exit(1)
def load_settings(self):
"""
Load settings for misowrap.
"""
settings_info, parsed_settings = \
misowrap_settings.load_misowrap_settings(self.settings_filename)
self.settings_info = settings_info
# Load basic settings about data
self.read_len = self.settings_info["settings"]["readlen"]
self.overhang_len = self.settings_info["settings"]["overhanglen"]
self.miso_bin_dir = \
utils.pathify(self.settings_info["settings"]["miso_bin_dir"])
self.miso_settings_filename = \
utils.pathify(self.settings_info["settings"]["miso_settings_filename"])
self.miso_events_dir = \
utils.pathify(self.settings_info["settings"]["miso_events_dir"])
self.miso_outdir = \
utils.pathify(self.settings_info["settings"]["miso_output_dir"])
# Load data-related parameters
self.bam_files = self.settings_info["data"]["bam_files"]
if "insert_lens_dir" in self.settings_info["data"]:
self.insert_lens_dir = \
utils.pathify(self.settings_info["data"]["insert_lens_dir"])
# Sample labels
self.sample_labels = self.settings_info["data"]["sample_labels"]
# Set output directories
self.comparisons_dir = os.path.join(self.output_dir, "comparisons")
self.comparison_groups = \
self.settings_info["data"]["comparison_groups"]
self.logs_outdir = os.path.join(self.output_dir, "misowrap_logs")
# Create necessary directories
utils.make_dir(self.logs_outdir)
if "cluster_type" in self.settings_info["settings"]:
self.use_cluster = True
self.cluster_type = \
self.settings_info["settings"]["cluster_type"]
self.chunk_jobs = \
self.settings_info["settings"]["chunk_jobs"]
if self.use_cluster:
print "Loading cluster information."
# Load cluster object if given a cluster type
self.load_cluster()
# Create a logger object
if self.logger_label is None:
self.logger_label = "misowrap"
else:
self.logger_label = "misowrap_%s" % (logger_label)
self.logger = utils.get_logger(self.logger_label, self.logs_outdir)
# Whether to prefilter MISO events
# Set general default settings
if "prefilter_miso" not in settings_info["settings"]:
# By default, set it so that MISO events are not
# prefiltered
settings_info["settings"]["prefilter_miso"] = False
self.prefilter_miso = \
self.settings_info["settings"]["prefilter_miso"]
# Load event types
self.load_event_types()
# Set path to MISO scripts
self.compare_miso_cmd = os.path.join(self.miso_bin_dir, "compare_miso")
self.summarize_miso_cmd = os.path.join(self.miso_bin_dir,
"summarize_miso")
self.run_events_cmd = os.path.join(self.miso_bin_dir, "miso")
self.pe_utils_cmd = os.path.join(self.miso_bin_dir, "pe_utils")
# Files related to gene tables
self.tables_dir = \
os.path.join(self.settings_info["pipeline-files"]["init_dir"],
"ucsc")
if not os.path.isdir(self.tables_dir):
print "Error: %s directory does not exist." \
%(self.tables_dir)
sys.exit(1)
self.const_exons_gff = os.path.join(self.tables_dir, "exons",
"const_exons",
"ensGene.const_exons.gff")
if not os.path.isfile(self.const_exons_gff):
print "Error: Const. exons GFF %s does not exist." \
%(self.const_exons_gff)
sys.exit(1)