def main():
args = parser.parse_args()
if args.output:
out_h = NamedTemporaryFile(dir=dirname(abspath(args.output)))
else:
out_h = stdout
blacklist = import_records([args.blacklist])
records = import_records(args.files)
(remaining, removals) = bind_filter(records, blacklist)
remaining = sort_records(remaining)
removals = sort_records(removals)
if args.removals:
for removal in removals:
(name,type,value) = removal
print >> stderr, name, type, value
else:
for (name,type,value) in remaining:
print >>out_h, name, type, value
if args.clean:
if len(removals) == 0:
if args.output:
rename(out_h.name, args.output)
out_h.delete = False
exit(0)
else:
exit(1)
else:
if args.output:
rename(out_h.name, args.output)
out_h.delete = False
exit(0)
def _generate_zip_file(self,
files_holder,
name_prefix='material',
name_suffix=None):
"""Generate a zip file containing the files passed.
:param files_holder: An iterable (or an iterable containing) object that
contains the files to be added in the zip file.
:param name_prefix: The prefix to the zip file name
:param name_suffix: The suffix to the zip file name
:return: The generated zip file.
"""
temp_file = NamedTemporaryFile(suffix='indico.tmp',
dir=config.TEMP_DIR)
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
self.used_filenames = set()
for item in self._iter_items(files_holder):
name = self._prepare_folder_structure(item)
self.used_filenames.add(name)
with item.storage.get_local_path(
item.storage_file_id) as filepath:
zip_handler.write(filepath.encode('utf-8'), name)
temp_file.delete = False
zip_file_name = '{}-{}.zip'.format(
name_prefix,
name_suffix) if name_suffix else '{}.zip'.format(name_prefix)
chmod_umask(temp_file.name)
return send_file(zip_file_name,
temp_file.name,
'application/zip',
inline=False)
def create(self):
"""Trigger the creation of a ZIP file containing the site."""
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=config.TEMP_DIR)
self._zip_file = ZipFile(temp_file.name, 'w', allowZip64=True)
with collect_static_files() as used_assets:
# create the home page html
html = self._create_home().encode('utf-8')
# Mathjax plugins can only be known in runtime
self._copy_folder(os.path.join(self._content_dir, 'static', 'dist', 'js', 'mathjax'),
os.path.join(self._static_dir, 'dist', 'js', 'mathjax'))
# Materials and additional pages
self._copy_all_material()
self._create_other_pages()
# Create index.html file (main page for the event)
index_path = os.path.join(self._content_dir, 'index.html')
self._zip_file.writestr(index_path, html)
self._write_generated_js()
# Copy static assets to ZIP file
self._copy_static_files(used_assets)
self._copy_plugin_files(used_assets)
if config.CUSTOMIZATION_DIR:
self._copy_customization_files(used_assets)
temp_file.delete = False
chmod_umask(temp_file.name)
self._zip_file.close()
return temp_file.name
def _generate_zip_file(self, files_holder, name_prefix='material', name_suffix=None):
"""Generate a zip file containing the files passed.
:param files_holder: An iterable (or an iterable containing) object that
contains the files to be added in the zip file.
:param name_prefix: The prifix to the zip file name
:param name_suffix: The suffix to the zip file name
:return: The generated zip file.
"""
from indico.util.tasks import delete_file
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=Config.getInstance().getTempDir())
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
self.used_filenames = set()
for item in self._iter_items(files_holder):
name = self._prepare_folder_structure(item)
self.used_filenames.add(name)
with item.storage.get_local_path(item.storage_file_id) as filepath:
zip_handler.write(filepath.encode('utf-8'), name)
# Delete the temporary file after some time. Even for a large file we don't
# need a higher delay since the webserver will keep it open anyway until it's
# done sending it to the client.
delete_file.apply_async(args=[temp_file.name], countdown=3600)
temp_file.delete = False
zip_file_name = '{}-{}.zip'.format(name_prefix, name_suffix) if name_suffix else '{}.zip'.format(name_prefix)
return send_file(zip_file_name, temp_file.name, 'application/zip', inline=False)
def _generate_zip_file(self, files_holder, name_prefix='material', name_suffix=None):
"""Generate a zip file containing the files passed.
:param files_holder: An iterable (or an iterable containing) object that
contains the files to be added in the zip file.
:param name_prefix: The prifix to the zip file name
:param name_suffix: The suffix to the zip file name
:return: The generated zip file.
"""
from indico.util.tasks import delete_file
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=Config.getInstance().getTempDir())
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
self.used_filenames = set()
for item in self._iter_items(files_holder):
name = self._prepare_folder_structure(item)
self.used_filenames.add(name)
with item.storage.get_local_path(item.storage_file_id) as filepath:
zip_handler.write(filepath.encode('utf-8'), name)
# Delete the temporary file after some time. Even for a large file we don't
# need a higher delay since the webserver will keep it open anyway until it's
# done sending it to the client.
delete_file.apply_async(args=[temp_file.name], countdown=3600)
temp_file.delete = False
zip_file_name = '{}-{}.zip'.format(name_prefix, name_suffix) if name_suffix else '{}.zip'.format(name_prefix)
return send_file(zip_file_name, temp_file.name, 'application/zip', inline=False)
def write_file(self, filename, data):
f = NamedTemporaryFile(prefix='hosted-py-tmp', dir=os.getcwd())
try:
f.write(data)
except:
traceback.print_exc()
raise
else:
f.delete = False
f.close()
os.rename(f.name, filename)
def _generate_zip_file(attachments, regform):
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=Config.getInstance().getTempDir())
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
for reg_attachments in attachments.itervalues():
for reg_attachment in reg_attachments:
name = _prepare_folder_structure(reg_attachment)
with reg_attachment.storage.get_local_path(reg_attachment.storage_file_id) as filepath:
zip_handler.write(filepath, name)
# Delete the temporary file after some time. Even for a large file we don't
# need a higher delay since the webserver will keep it open anyway until it's
# done sending it to the client.
delete_file.apply_async(args=[temp_file.name], countdown=3600)
temp_file.delete = False
return send_file('attachments-{}.zip'.format(regform.id), temp_file.name, 'application/zip', inline=False)
def _generate_zip_file(attachments, regform):
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=Config.getInstance().getTempDir())
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
for reg_attachments in attachments.itervalues():
for reg_attachment in reg_attachments:
name = _prepare_folder_structure(reg_attachment)
with reg_attachment.storage.get_local_path(reg_attachment.storage_file_id) as filepath:
zip_handler.write(filepath, name)
# Delete the temporary file after some time. Even for a large file we don't
# need a higher delay since the webserver will keep it open anyway until it's
# done sending it to the client.
delete_file.apply_async(args=[temp_file.name], countdown=3600)
temp_file.delete = False
return send_file('attachments-{}.zip'.format(regform.id), temp_file.name, 'application/zip', inline=False)
def scratch_cached(self, filename, generator):
cached = os.path.join(os.environ['SCRATCH'], filename)
if not os.path.exists(cached):
f = NamedTemporaryFile(prefix='scratch-cached-tmp', dir=os.environ['SCRATCH'])
try:
generator(f)
except:
raise
else:
f.delete = False
f.close()
os.rename(f.name, cached)
if os.path.exists(filename):
try:
os.unlink(filename)
except:
pass
os.symlink(cached, filename)
def _generate_zip_file(self, files_holder, name_prefix='material', name_suffix=None):
"""Generate a zip file containing the files passed.
:param files_holder: An iterable (or an iterable containing) object that
contains the files to be added in the zip file.
:param name_prefix: The prefix to the zip file name
:param name_suffix: The suffix to the zip file name
:return: The generated zip file.
"""
temp_file = NamedTemporaryFile(suffix='indico.tmp', dir=config.TEMP_DIR)
with ZipFile(temp_file.name, 'w', allowZip64=True) as zip_handler:
self.used_filenames = set()
for item in self._iter_items(files_holder):
name = self._prepare_folder_structure(item)
self.used_filenames.add(name)
with item.storage.get_local_path(item.storage_file_id) as filepath:
zip_handler.write(filepath.encode('utf-8'), name)
temp_file.delete = False
zip_file_name = '{}-{}.zip'.format(name_prefix, name_suffix) if name_suffix else '{}.zip'.format(name_prefix)
chmod_umask(temp_file.name)
return send_file(zip_file_name, temp_file.name, 'application/zip', inline=False)
# let outf get auto-deleted if we made no changes
outf.close()
if os.name=='nt':
os.unlink(outf.name)
else:
if isinstance(args.inplace, basestring):
ext = args.inplace
if os.path.exists(inf.name + ext):
p.error("can't make backup of %s: %s already exists" % (inf.name, inf.name + ext))
try:
os.rename(inf.name, inf.name + ext)
except OSError as e:
p.error("can't rename %s to %s: %s" % (inf.name, inf.name + ext, str(e)))
# don't mark output file with delete=False until all the preceding steps have succeeded
outf.delete = False
outf.close()
shutil.copymode(inf.name + ext, outf.name)
else:
outf.delete = False
outf.close()
shutil.copymode(inf.name, outf.name)
# rename won't work on Windows if destination exists
os.unlink(inf.name)
os.rename(outf.name, inf.name)
if not args.inplace:
outf.close()
if not args.no_exit_codes:
if os.name == 'nt':
os.unlink(outf.name)
else:
if isinstance(args.inplace, basestring):
ext = args.inplace
if os.path.exists(inf.name + ext):
p.error("can't make backup of %s: %s already exists" %
(inf.name, inf.name + ext))
try:
os.rename(inf.name, inf.name + ext)
except OSError as e:
p.error("can't rename %s to %s: %s" %
(inf.name, inf.name + ext, str(e)))
# don't mark output file with delete=False until all the preceding steps have succeeded
outf.delete = False
outf.close()
shutil.copymode(inf.name + ext, outf.name)
else:
outf.delete = False
outf.close()
shutil.copymode(inf.name, outf.name)
# rename won't work on Windows if destination exists
os.unlink(inf.name)
os.rename(outf.name, inf.name)
if not args.inplace:
outf.close()
if not args.no_exit_codes:
def main():
"""
1: Collect Tx from one or more species that are within at least some r value of similarity to
a provided example Tx or a submitted hypothetical expression vector.
2: Use GTFs, BEDtools, and genome FASTAs to extract the upstream flanking sequences into a new FASTA
for use in motif discovery.
"""
desc = """(1) Collect Tx from one or more species that are within
at least some r value of similarity to a provided example Tx or a
submitted hypothetical expression vector. (2) Use GTFs, BEDtools, and
genome FASTAs to extract the upstream flanking sequences into a new
FASTA for use in motif discovery."""
parser = argparse.ArgumentParser(description=desc)
FileType = argparse.FileType
logger = logging.getLogger(sys.argv[0].split('/')[-1])
parser.add_argument('--expn-path', type=str, required=True,
help="""Path to expression table file. \n(default: %(default)s)""")
parser.add_argument('--tx-name', type=str, required=True,
help="""Name of the Tx you want to use as a model. (default: %(default)s)""")
parser.add_argument('--pearson-filter-type', type=str, default='>=', choices=['>=','<='],
help="""Use >= to find similar expn profiles or <= to find opposite profiles. (default: %(default)s)""")
parser.add_argument('--pearson-filter-thresh', type=float, default=0.7,
help="""Set the threshold of the Pearson r value for the filter. (default: %(default)s)""")
parser.add_argument('--pval-filter-thresh', type=float, default=0.05,
help="""Set the upper threshold for the p-value of the Pearson r values to keep. (default: %(default)s)""")
parser.add_argument('--tx-name-header', type=str, required=True,
help="""The text of the header in the expn table where tx names are stored. (default: %(default)s)""")
parser.add_argument('--cond-headers', type=str, required=True, nargs='+',
help="""A list of the text of the headers in the expn table where the values for each condition are stored (--cond-headers cond1 cond2 ...). (default: %(default)s)""")
parser.add_argument('--manual-headers', type=str, required=False, nargs='?',
help="""If the expn table does not have headers, provide a list of ordered names for them here. (default: %(default)s)""")
parser.add_argument('--gtf', type=str, required=True,
help="""The path to the gtf file that you want to use for your annotation. (default: %(default)s)""")
parser.add_argument('--gtf-index', type=str, required=True,
help="""The path to the gtf index file generated from "gtf_to_genes". (default: %(default)s)""")
parser.add_argument('--genome-fastas', type=str, required=True, nargs='+',
help="""A list of paths to genomic fasta files or directories where they are stored. (default: %(default)s)""")
parser.add_argument('--flank-len', type=int, default=2000,
help="""The length in bp that should be harvested from the 5' end of the tx. (default: %(default)s)""")
parser.add_argument('--out-dir', type=str, default='.',
help="""A path to a directory where you would like the output files to be stored. (default: %(default)s)""")
parser.add_argument('--dump-megafasta', action='store_true',
help="""Save concatonated fasta file for debugging. (default: %(default)s)""")
parser.add_argument('--dump-stats', action='store_true',
help="""Print a list of Tx/gene names and the r- p-values that passed the filter and exit without getting fastas. (default: %(default)s)""")
args = parser.parse_args()
# tmp files will be stored here
tmp_files = Bag()
# 1: Use a correlation filter to pull out any Tx that is sufficiently similar to the model Tx
vectDict = mangle_expn_vectors(expnPath=args.expn_path,txNameHeader=args.tx_name_header,condHeaders=args.cond_headers,manualHeaders=args.manual_headers)
filterFunc = eval("lambda x: x %s %f" % (args.pearson_filter_type, args.pearson_filter_thresh))
filterDict = pearsonExpnFilter(modelVector=vectDict[args.tx_name], targetVectors=vectDict, filterFunc=filterFunc)
# remove vectors whose r's pVal is not significant (<=0.05)
sigVectors = {}
for key in filterDict:
if key[1] <= args.pval_filter_thresh:
sigVectors[key] = filterDict[key]
matchVectors = sigVectors
## Impose a distance filter to further refine the gene set
## incorperating magnitudes of the absolute levels of gene expression
## set the boundries of acceptable deviation for the target gene mean expression
## mangitude by bootstrapping. The metric for comparison will be the average of
## the differences of each point in remaining vectors against the target
## vector.
## 1) calc the metrics for each remaining gene's vector
## PS: numpy rocks.
##avgDists = {}
##for key in sigVectors:
##avgDist_i = np.mean(np.subtract(vectDict[args.tx_name],
##sigVectors[key]))
##avgDists[key] = avgDist_i
### 2) bootstrap that bitch and give me a stdErr!
##medianEst,stdErrEst,lo95,hi95 = basic_bootstrap_est(avgDists.values())
### 3) recover keys that fall within +/- 1 SE
##matchVectors = {}
##for key in avgDists:
##avgDist = avgDists[key]
##if (avgDist >= -stdErrEst) and (avgDist <= stdErrEst):
##matchVectors[key] = sigVectors[key]
# Sort txList so that the highest r values are at the top
# and save vectors and this info out to file
txList = sorted(matchVectors.keys(),key=lambda x: x[0], reverse=True)
sortedTxListFile = NamedTemporaryFile(mode='w+t',prefix='txExpnVectFilteredBy_r.',suffix=".tsv",delete=False)
for row in txList:
if args.dump_stats:
sys.stdout.write('%s\t%s\n' % ('\t'.join(map(str,row)),'\t'.join(map(str,matchVectors[row]))))
else:
sortedTxListFile.write('%s\t%s\n' % ('\t'.join(map(str,row)),'\t'.join(map(str,matchVectors[row]))))
if args.dump_stats:
sortedTxListFile.close()
exit(0)
tmp_files['sortedTxListFile'] = sortedTxListFile
sortedTxListFile.close()
g2gObj = gtf_to_genes.get_indexed_genes_matching_gtf_file_name(index_file_name=args.gtf_index, logger=logger, regex_str=args.gtf)[-1]
txDict = filter_GTF_4_Tx(txList=[x[2] for x in txList],g2gObj=g2gObj)
tmp_files['txBedFile'] = convert_2_bed(txDict=txDict)
# 2: Use GTFs, BEDtools, and genome FASTAs to extract the upstream flanking sequences into a new FASTA
fastaRecLengths,fastaSeqs = fastaRec_length_indexer(fastaFiles=args.genome_fastas)
tmpFastaRecLengthFile = NamedTemporaryFile(mode='w+b',prefix='tmpFastaRecLengthFile.',suffix=".txt")
for seqRec in fastaRecLengths:
tmpFastaRecLengthFile.write("%s\t%s\n" % (seqRec,fastaRecLengths[seqRec]))
tmpFastaRecLengthFile.flush()
# TODO: concatonate fasta files
megaFastaFile = NamedTemporaryFile(mode='w+b',prefix='tmpMegaFastaFile.',suffix=".fas")
for fasta in fastaSeqs:
megaFastaFile.write('>%s\n%s\n' % (fasta,fastaSeqs[fasta]))
megaFastaFile.flush()
tmp_files['flankBed'] = get_fastas(txBed=tmp_files.txBedFile.name,genomeFasta=megaFastaFile.name,lenIndex=tmpFastaRecLengthFile.name,lenFlanks=args.flank_len)
# CLEAN UP:
# TODO: Close all tmp_files, and move to args.outDir
mkdirp(args.out_dir)
for f in tmp_files:
try:
tmp_files[f].delete = False
except AttributeError:
pass
try:
tmp_files[f].close()
except AttributeError:
pass
# ['sortedTxListFile', 'flankBed', 'txBedFile', 'flankFasta']
sortedTxListFile = "%s/sortedTxList.tsv" % (args.out_dir)
flankBed = "%s/flankBed.bed" % (args.out_dir)
txBedFile = "%s/txBed.bed" % (args.out_dir)
flankFasta = "%s/flankFasta.fas" % (args.out_dir)
shutil.move(tmp_files.sortedTxListFile.name, sortedTxListFile)
os.chmod(sortedTxListFile,0775)
tmp_files.flankBed.saveas(flankBed)
os.chmod(flankBed,0775)
shutil.move(tmp_files.txBedFile.name, txBedFile)
os.chmod(txBedFile,0775)
shutil.move(tmp_files.flankBed.seqfn, flankFasta)
os.chmod(flankFasta,0775)
if args.dump_megafasta:
megaFasta = "%s/megaFasta.fas" % (args.out_dir)
megaFastaFile.delete = False
megaFastaFile.close()
shutil.move(megaFastaFile.name, megaFasta)
os.chmod(megaFasta,0775)
