tree = ElementTree.parse(filename).getroot()
self.set_elements(tree)
+def load_pipeline_run_xml(pathname):
+ """
+ Load and instantiate a Pipeline run from a run xml file
+
+ :Parameters:
+ - `pathname` : location of an run xml file
+
+ :Returns: initialized PipelineRun object
+ """
+ tree = ElementTree.parse(pathname).getroot()
+ run = PipelineRun(xml=tree)
+ return run
+
def get_runs(runfolder):
"""
Search through a run folder for all the various sub component runs
p.gerald = g
runs.append(p)
except IOError, e:
- print "Ignoring", str(e)
+ logging.error("Ignoring " + str(e))
datadir = os.path.join(runfolder, 'Data')
for firecrest_pathname in glob(os.path.join(datadir,"*Firecrest*")):
logging.info('Found firecrest in ' + datadir)
image_analysis = firecrest.firecrest(firecrest_pathname)
- scan_post_image_analysis(runs, runfolder, image_analysis, firecrest_pathname)
+ if image_analysis is None:
+ logging.warn(
+ "%s is an empty or invalid firecrest directory" % (firecrest_pathname,)
+ )
+ else:
+ scan_post_image_analysis(
+ runs, runfolder, image_analysis, firecrest_pathname
+ )
# scan for IPAR directories
for ipar_pathname in glob(os.path.join(datadir,"IPAR_*")):
logging.info('Found ipar directories in ' + datadir)
image_analysis = ipar.ipar(ipar_pathname)
- scan_post_image_analysis(runs, runfolder, image_analysis, ipar_pathname)
+ if image_analysis is None:
+ logging.warn(
+ "%s is an empty or invalid IPAR directory" %(ipar_pathname,)
+ )
+ else:
+ scan_post_image_analysis(
+ runs, runfolder, image_analysis, ipar_pathname
+ )
return runs
+def get_specific_run(gerald_dir):
+ """
+ Given a gerald directory, construct a PipelineRun out of its parents
+
+ Basically this allows specifying a particular run instead of the previous
+ get_runs which scans a runfolder for various combinations of
+ firecrest/ipar/bustard/gerald runs.
+ """
+ from htsworkflow.pipelines import firecrest
+ from htsworkflow.pipelines import ipar
+ from htsworkflow.pipelines import bustard
+ from htsworkflow.pipelines import gerald
+
+ bustard_dir = os.path.abspath(os.path.join(gerald_dir, '..'))
+ image_dir = os.path.abspath(os.path.join(gerald_dir, '..', '..'))
+
+ runfolder_dir = os.path.abspath(os.path.join(image_dir, '..','..'))
+
+ logging.info('--- use-run detected options ---')
+ logging.info('runfolder: %s' % (runfolder_dir,))
+ logging.info('image_dir: %s' % (image_dir,))
+ logging.info('bustard_dir: %s' % (bustard_dir,))
+ logging.info('gerald_dir: %s' % (gerald_dir,))
+
+ # find our processed image dir
+ image_run = None
+ # split into parent, and leaf directory
+ # leaf directory should be an IPAR or firecrest directory
+ data_dir, short_image_dir = os.path.split(image_dir)
+ logging.info('data_dir: %s' % (data_dir,))
+ logging.info('short_iamge_dir: %s' %(short_image_dir,))
+
+ # guess which type of image processing directory we have by looking
+ # in the leaf directory name
+ if re.search('Firecrest', short_image_dir, re.IGNORECASE) is not None:
+ image_run = firecrest.firecrest(image_dir)
+ elif re.search('IPAR', short_image_dir, re.IGNORECASE) is not None:
+ image_run = ipar.ipar(image_dir)
+ # if we din't find a run, report the error and return
+ if image_run is None:
+ msg = '%s does not contain an image processing step' % (image_dir,)
+ logging.error(msg)
+ return None
+
+ # find our base calling
+ base_calling_run = bustard.bustard(bustard_dir)
+ if base_calling_run is None:
+ logging.error('%s does not contain a bustard run' % (bustard_dir,))
+ return None
+
+ # find alignments
+ gerald_run = gerald.gerald(gerald_dir)
+ if gerald_run is None:
+ logging.error('%s does not contain a gerald run' % (gerald_dir,))
+ return None
+
+ p = PipelineRun(runfolder_dir)
+ p.image_analysis = image_run
+ p.bustard = base_calling_run
+ p.gerald = gerald_run
+
+ logging.info('Constructed PipelineRun from %s' % (gerald_dir,))
+ return p
def extract_run_parameters(runs):
"""
for run in runs:
run.save()
-def summarize_mapped_reads(mapped_reads):
+def summarize_mapped_reads(genome_map, mapped_reads):
"""
Summarize per chromosome reads into a genome count
But handle spike-in/contamination symlinks seperately.
genome = 'unknown'
for k, v in mapped_reads.items():
path, k = os.path.split(k)
- if len(path) > 0:
+ if len(path) > 0 and not genome_map.has_key(path):
genome = path
genome_reads += v
else:
report.append('Repeat (0,1,2 mismatches) %d %d %d' % \
(mc['R0'], mc['R1'], mc['R2']))
report.append("Mapped Reads")
- mapped_reads = summarize_mapped_reads(eland_result.mapped_reads)
+ mapped_reads = summarize_mapped_reads(eland_result.genome_map, eland_result.mapped_reads)
for name, counts in mapped_reads.items():
report.append(" %s: %d" % (name, counts))
report.append('')
# tar score files
score_files = []
- for f in os.listdir(g.pathname):
+
+ # check for g.pathname/Temp a new feature of 1.1rc1
+ scores_path = g.pathname
+ scores_path_temp = os.path.join(scores_path, 'Temp')
+ if os.path.isdir(scores_path_temp):
+ scores_path = scores_path_temp
+
+ # hopefully we have a directory that contains s_*_score files
+ for f in os.listdir(scores_path):
if re.match('.*_score.txt', f):
score_files.append(f)
bzip_cmd = [ 'bzip2', '-9', '-c' ]
tar_dest_name =os.path.join(cycle_dir, 'scores.tar.bz2')
tar_dest = open(tar_dest_name, 'w')
- logging.info("Compressing score files in %s" % (g.pathname,))
+ logging.info("Compressing score files from %s" % (scores_path,))
logging.info("Running tar: " + " ".join(tar_cmd[:10]))
logging.info("Running bzip2: " + " ".join(bzip_cmd))
logging.info("Writing to %s" %(tar_dest_name))
- tar = subprocess.Popen(tar_cmd, stdout=subprocess.PIPE, shell=False, cwd=g.pathname)
+ tar = subprocess.Popen(tar_cmd, stdout=subprocess.PIPE, shell=False,
+ cwd=scores_path)
bzip = subprocess.Popen(bzip_cmd, stdin=tar.stdout, stdout=tar_dest)
tar.wait()
source_name = eland_lane.pathname
path, name = os.path.split(eland_lane.pathname)
dest_name = os.path.join(cycle_dir, name)
+ logging.info("Saving eland file %s to %s" % \
+ (source_name, dest_name))
+
if is_compressed(name):
logging.info('Already compressed, Saving to %s' % (dest_name, ))
shutil.copy(source_name, dest_name)