def test_repository(self):
"""
test that the repository object gets made
"""
repository = StageRepository(self.config)
self.assertTrue(isinstance(repository, StageRepository))
def setUp(self):
self.config_file = os.path.join("test", STAGENAME,
"test_" + STAGENAME + ".yaml")
with open(self.config_file) as in_handle:
self.config = yaml.load(in_handle)
self.repository = StageRepository(self.config)
self.out_dir = os.path.join(self.config["dir"]["results"], "test",
STAGENAME)
safe_makedir(self.out_dir)
def test_custom_plugins(self):
"""
test that the custom plugin directory is working
"""
plugin_file = self._make_test_class()
plugin_dir = os.path.dirname(plugin_file.name)
repo = StageRepository({"dir": {"plugins": plugin_dir}})
testplugin = repo["test_plugin"](self.config)
self.assertTrue(testplugin("Test") == "Test")
def main(config_file):
with open(config_file) as in_handle:
config = yaml.load(in_handle)
# make the needed directories
map(safe_makedir, config["dir"].values())
# specific for project
input_dir = config["dir"]["data"]
logger.info("Loading files from %s" % (input_dir))
input_files = list(locate("*.fq", input_dir))
input_files += list(locate("*.fastq", input_dir))
logger.info("Input files: %s" % (input_files))
results_dir = config["dir"]["results"]
safe_makedir(results_dir)
# make the stage repository
repository = StageRepository(config)
logger.info("Stages found: %s" % (repository.plugins))
if config.get("test_pipeline", False):
logger.info("Running a test pipeline on a subset of the reads.")
results_dir = os.path.join(results_dir, "test_pipeline")
config["dir"]["results"] = results_dir
safe_makedir(results_dir)
curr_files = map(make_test, input_files, [config] * len(input_files))
logger.info("Converted %s to %s. " % (input_files, curr_files))
else:
curr_files = input_files
logger.info("Running RNASeq alignment pipeline on %s." % (curr_files))
for stage in config["run"]:
if stage == "fastqc":
logger.info("Running fastqc on %s." % (curr_files))
stage_runner = FastQC(config)
view.map(stage_runner, curr_files)
if stage == "cutadapt":
curr_files = combine_pairs(curr_files)
logger.info("Running cutadapt on %s." % (curr_files))
stage_runner = Cutadapt(config)
curr_files = view.map(stage_runner, curr_files)
if stage == "tophat":
logger.info("Running Tophat on %s." % (curr_files))
#tophat = repository["tophat"](config)
tophat = Tophat(config)
tophat_outputs = view.map(tophat, curr_files)
bamfiles = view.map(sam.sam2bam, tophat_outputs)
bamsort = view.map(sam.bamsort, bamfiles)
view.map(sam.bamindex, bamsort)
final_bamfiles = bamsort
curr_files = tophat_outputs
if stage == "disambiguate":
logger.info("Disambiguating %s." % (curr_files))
disambiguate = repository[stage](config)
view.map(disambiguate, curr_files)
if stage == "htseq-count":
logger.info("Running htseq-count on %s." % (bamfiles))
name_sorted = view.map(sam.bam_name_sort, bamfiles)
curr_files = view.map(sam.bam2sam, name_sorted)
htseq_args = zip(*product(curr_files, [config], [stage]))
htseq_outputs = view.map(htseq_count.run_with_config, *htseq_args)
htseq_count.combine_counts(htseq_outputs)
if stage == "rnaseq_metrics":
logger.info("Calculating RNASeq metrics on %s." % (curr_files))
#coverage = repository[stage](config)
coverage = RNASeqMetrics(config)
view.map(coverage, curr_files)
if stage == "rseqc":
logger.info("Running rseqc on %s." % (curr_files))
#rseq_args = zip(*product(curr_files, [config]))
rseq_args = zip(*product(final_bamfiles, [config]))
view.map(rseqc.bam_stat, *rseq_args)
view.map(rseqc.genebody_coverage, *rseq_args)
view.map(rseqc.junction_annotation, *rseq_args)
view.map(rseqc.junction_saturation, *rseq_args)
RPKM_args = zip(*product(final_bamfiles, [config]))
RPKM_count_out = view.map(rseqc.RPKM_count, *RPKM_args)
RPKM_count_fixed = view.map(rseqc.fix_RPKM_count_file,
RPKM_count_out)
"""
annotate_args = zip(*product(RPKM_count_fixed,
["gene_id"],
["ensembl_gene_id"],
["human"]))
view.map(annotate.annotate_table_with_biomart,
*annotate_args)
"""
view.map(rseqc.RPKM_saturation, *rseq_args)
curr_files = tophat_outputs
# end gracefully
stop_cluster()
def main(config_file):
with open(config_file) as in_handle:
config = yaml.load(in_handle)
# make the needed directories
map(safe_makedir, config["dir"].values())
# specific for project
input_dir = config["input_dir"]
logger.info("Loading files from %s" % (input_dir))
input_files = list(
locate("*.sam", os.path.join(input_dir, "tophat_control")))
input_files += list(
locate("*.sam", os.path.join(input_dir, "tophat_exposed")))
print input_files
input_files = [x for x in input_files if "accepted" not in x]
input_files = [x for x in input_files if "innerdist_estimate" not in x]
logger.info("Input files: %s" % (input_files))
results_dir = config["dir"]["results"]
safe_makedir(results_dir)
# make the stage repository
repository = StageRepository(config)
logger.info("Stages found: %s" % (repository.plugins))
curr_files = input_files
logger.info("Running quantitation on %s." % (curr_files))
for stage in config["run"]:
if stage == "htseq-count":
logger.info("Running htseq-count on %s." % (curr_files))
htseq_args = zip(*product(curr_files, [config], [stage]))
htseq_outputs = view.map(htseq_count.run_with_config, *htseq_args)
htseq_count.combine_counts(htseq_outputs)
if stage == "rnaseq_metrics":
logger.info("Calculating RNASeq metrics on %s." % (curr_files))
#coverage = repository[stage](config)
curr_files = view.map(sam.bam2sam, curr_files)
coverage = RNASeqMetrics(config)
view.map(coverage, curr_files)
if stage == "rseqc":
logger.info("Running rseqc on %s." % (curr_files))
rseq_args = zip(*product(curr_files, [config]))
view.map(rseqc.bam_stat, *rseq_args)
view.map(rseqc.genebody_coverage, *rseq_args)
view.map(rseqc.junction_annotation, *rseq_args)
view.map(rseqc.junction_saturation, *rseq_args)
RPKM_args = zip(*product(final_bamfiles, [config]))
RPKM_count_out = view.map(rseqc.RPKM_count, *RPKM_args)
RPKM_count_fixed = view.map(rseqc.fix_RPKM_count_file,
RPKM_count_out)
"""
annotate_args = zip(*product(RPKM_count_fixed,
["gene_id"],
["ensembl_gene_id"],
["human"]))
view.map(annotate.annotate_table_with_biomart,
*annotate_args)
"""
view.map(rseqc.RPKM_saturation, *rseq_args)
# end gracefully
stop_cluster()