def runTest(self):
outdir = '/tmp/gMiner/'
if not os.path.exists(outdir): os.mkdir(outdir)
# Cross-correlation #
files = run(
track1 = track_collections['Scores'][1]['path_sql'],
track1_name = 'Scores 1',
track1_chrs = small_chr_file,
track2 = track_collections['Scores'][2]['path_sql'],
track2_name = 'Scores 2',
track2_chrs = small_chr_file,
operation_type = 'plot',
plot = 'correlation',
output_location = outdir,
)
# Scatter plot #
files = run(
track1 = track_collections['Scores'][1]['path_sql'],
track1_name = 'Scores 1',
track1_chrs = small_chr_file,
track2 = track_collections['Scores'][2]['path_sql'],
track2_name = 'Scores 2',
track2_chrs = small_chr_file,
track3 = track_collections['Validation'][1]['path_sql'],
track3_name = 'Features 1',
track3_chrs = small_chr_file,
operation_type = 'plot',
plot = 'scatter',
output_location = outdir,
)
def runTest(self):
outdir = '/tmp/gMiner/'
if not os.path.exists(outdir): os.mkdir(outdir)
# Simple boxplot #
files = run(
track1 = track_collections['Yeast']['All genes']['path_sql'],
track1_name = 'All genes from SGD',
track2 = track_collections['Yeast']['Ribi genes']['path_sql'],
track2_name = 'Ribosome genesis',
track3 = track_collections['Yeast']['RP genes']['path_sql'],
track3_name = 'Ribosome proteins',
operation_type = 'desc_stat',
characteristic = 'length',
output_location = outdir,
)
# Track selection #
files = run(
track1 = track_collections['Validation'][2]['path_sql'],
track1_name = 'Validation track two',
operation_type = 'desc_stat',
characteristic = 'base_coverage',
selected_regions = track_collections['Validation'][3]['path_sql'],
per_chromosome = True,
compare_parents = True,
output_location = outdir,
)
# Result should be 120 / 150 or 80%
# Field test via filter #
files = run(
track1 = track_collections['Yeast']['All genes']['path_sql'],
track1_name = 'S. cer refseq genes',
track2 = track_collections['Yeast']['RP genes']['path_sql'],
track2_name = 'RP genes',
operation_type = 'desc_stat',
characteristic = 'number_of_features',
selected_regions = track_collections['Yeast']['Ribi genes']['path_sql'],
per_chromosome = True,
compare_parents = False,
output_location = outdir,
)
# Permission test #
if os.path.exists('/scratch/genomic/tracks/locked.sql'):
files = run(
track1 = '/scratch/genomic/tracks/locked.sql',
track1_name = 'Permission test',
selected_regions = '/scratch/genomic/tracks/locked.sql',
operation_type = 'desc_stat',
characteristic = 'number_of_features',
output_location = outdir,
)
def generate(result_path='/tmp/gMiner/'):
if not os.path.isdir(result_path):
raise Exception("The result location specified is not a directory")
max_count = 2*2*2*2*4
max_count -= max_count/4
max_count *= len(collections)
count = 1
print gm_terminal_colors['bldylw'] + " Writing graphs in: " + result_path + gm_terminal_colors['end']
for col_name, col in collections.items():
for b_many in [True, False]:
for b_sel in [True, False]:
for b_comp in [True, False]:
for b_chr in [True, False]:
for chara in chara_dict.keys():
# Preparation #
if not b_sel and b_comp: continue
name = col_name + '_'
name = name + name_dict[b_comp][b_chr][b_many]
name = name + chara_dict[chara]
if not b_sel: name += '_NoSel'
print gm_terminal_colors['txtylw'] + "Graph " + str(count) + " out of " + str(max_count) + \
" " + gm_terminal_colors['txtpur'] + name + gm_terminal_colors['end']
count += 1
# Making the request #
request = {}
request['gm_encoding'] = 'image/png'
request['operation_type'] = 'desc_stat'
request['characteristic'] = chara
request['output_location'] = result_path
request['output_name'] = name
if b_many:
request['track1'] = col['track_set']['many'][1]['path_sql']
request['track1_name'] = col['track_name_fn'](col['track_set']['many'][1]['name'])
request['track2'] = col['track_set']['many'][2]['path_sql']
request['track2_name'] = col['track_name_fn'](col['track_set']['many'][2]['name'])
request['track3'] = col['track_set']['many'][3]['path_sql']
request['track3_name'] = col['track_name_fn'](col['track_set']['many'][3]['name'])
else:
request['track1'] = col['track_set']['single']['path_sql']
request['track1_name'] = col['track_name_fn'](col['track_set']['single']['name'])
if b_sel:
request['selected_regions'] = col['request_selection_string']
if b_chr:
request['per_chromosome'] = 'True'
if b_comp:
request['compare_parents'] = 'True'
# Executing the request #
run(**request)
def runTest(self):
path = track_collections['Validation'][1]['path']
files = run(
track1 = path,
track1_name = 'Validation track one',
operation_type = 'genomic_manip',
manipulation = 'merge',
output_location = tempfile.gettempdir(),
)
os.remove(files[0])
def runTest(self):
sql_path = track_collections['Validation'][2]['path_sql']
bed_path = track_collections['Validation'][2]['path']
files = run(
track1 = bed_path,
track1_name = 'Validation track two',
track1_chrs = yeast_chr_file,
operation_type = 'genomic_manip',
manipulation = 'bool_not',
output_location = tempfile.gettempdir(),
)
with load(sql_path, chrmeta=yeast_chr_file, readonly=True) as sql:
with load(files[0]) as bed:
self.assertEqual(sql.chrmeta, bed.chrmeta)
os.remove(files[0])
def main(argv = None):
if argv is None:
argv = sys.argv[1:]
try:
if len(argv) != 1:
raise Usage("run_gminer.py takes exactly 1 argument.")
pickle_file = argv[0]
with open(pickle_file,'r') as f:
job = pickle.load(f)
job['job_output'] = gMiner.run(**job)
with open(pickle_file,'w') as f:
pickle.dump(job,f)
sys.exit(0)
except Usage, err:
print >>sys.stderr, err.msg
print >>sys.stderr, usage
sys.exit(2)
def post_process(**kwargs):
try:
# Get a job from the list #
global jobs
job = jobs.pop(0)
# Load the form #
request = json.loads(job['data'])
# Get the output location #
request['output_location'] = job.pop('output_location')
# Format the request #
if request.has_key('compare_parents' ): request['compare_parents' ] = bool(request['compare_parents' ])
if request.has_key('per_chromosome' ): request['per_chromosome' ] = bool(request['per_chromosome' ])
if request.has_key('filter'):
request['selected_regions'] = request['filter'][0]['path']
request.pop('filter')
if request.has_key('ntracks'):
request.update(dict([('track' + str(i+1), v['path']) for i,v in enumerate(request['ntracks'])]))
request.update(dict([('track' + str(i+1) + '_name', v.get('name', 'Unamed')) for i,v in enumerate(request['ntracks'])]))
request.pop('ntracks')
# Unicode filtering #
request = dict([(k.encode('ascii'),v) for k,v in request.items()])
# Run the request #
files = gMiner.run(**request)
# Format the output #
result = {'files': [dict([('path',p),('type',p.split('.')[-1])]) for p in files]}
# Report success #
print '\033[1;33m[' + time.asctime() + ']\033[0m \033[42m' + files[0] + '\033[0m'
except Exception as err:
traceback.print_exc()
print '\033[1;33m[' + time.asctime() + ']\033[0m \033[41m' + str(err) + '\033[0m'
with warnings.catch_warnings():
warnings.simplefilter("ignore")
result = {'type':'error', 'html':cgitb.html(sys.exc_info()), 'msg': str(err)}
finally:
result = locals().get('result', '')
connection = httplib2.Http()
body = urllib.urlencode({'id':'job', 'action':'gfeatresponse', 'job_id':job.get('job_id', -1), 'data':json.dumps(result)})
headers = {'content-type': 'application/x-www-form-urlencoded'}
address = job['callback_url']
response, content = connection.request(address, "POST", body=body, headers=headers)
import gMiner
files = gMiner.run(
track1 = '/scratch/genomic/tracks/ribosome_proteins.sql',
track1_name = 'RP genes (SGD)',
track2 = '/scratch/genomic/tracks/ribosome_genesis.sql',
track2_name = 'Ribi genes (SGD)',
operation_type = 'desc_stat',
characteristic = 'base_coverage',
selected_regions = 'chr2:0:300000;chr5:0:200000',
compare_parents = 'False',
per_chromosome = 'True',
wanted_chromosomes = 'chr1;chr2;chr5;chr6;chr7;chr8;chr9;chr10;chr11;chr12;' + \
'chr13;chr14;chr15;chr16;chr17;chr18;chr19;chr20;chr21;' + \
'chr22;chrX;chrY',
output_location = '/tmp/',
)
# Modules #
import sys, tempfile
# gMiner #
import gMiner
from gMiner.gm_constants import *
# Parser #
def parse_args(args):
option = ''
for arg in args:
if arg.startswith("--"): option = arg[2:]
else:
if not option:
print "Bad options"
sys.exit()
else:
yield (option, arg)
# Make the request #
request = dict([(x[0], x[1]) for x in parse_args(sys.argv[1:])])
# Run it #
print gMiner.run(**request)
import gMiner
files = gMiner.run(
track1 = '/scratch/genomic/tracks/all_yeast_genes.sql',
track1_name = 'S. cer. genes (SGD)',
operation_type = 'desc_stat',
characteristic = 'number_of_features',
per_chromosome = 'True',
compare_parents = 'False',
output_location = '/tmp/',
)
def post_process(**kwargs):
try:
# Get a job from the list #
global jobs
job = jobs.pop(0)
# Load the form #
request = json.loads(job['data'])
# Get the output location #
request['output_location'] = job.pop('output_location')
# Format the request #
if request.has_key('compare_parents'):
request['compare_parents'] = bool(request['compare_parents'])
if request.has_key('per_chromosome'):
request['per_chromosome'] = bool(request['per_chromosome'])
if request.has_key('filter'):
request['selected_regions'] = request['filter'][0]['path']
request.pop('filter')
if request.has_key('ntracks'):
request.update(
dict([('track' + str(i + 1), v['path'])
for i, v in enumerate(request['ntracks'])]))
request.update(
dict([('track' + str(i + 1) + '_name', v.get('name', 'Unamed'))
for i, v in enumerate(request['ntracks'])]))
request.pop('ntracks')
# Unicode filtering #
request = dict([(k.encode('ascii'), v) for k, v in request.items()])
# Run the request #
files = gMiner.run(**request)
# Format the output #
result = {
'files':
[dict([('path', p), ('type', p.split('.')[-1])]) for p in files]
}
# Report success #
print '\033[1;33m[' + time.asctime(
) + ']\033[0m \033[42m' + files[0] + '\033[0m'
except Exception as err:
traceback.print_exc()
print '\033[1;33m[' + time.asctime() + ']\033[0m \033[41m' + str(
err) + '\033[0m'
with warnings.catch_warnings():
warnings.simplefilter("ignore")
result = {
'type': 'error',
'html': cgitb.html(sys.exc_info()),
'msg': str(err)
}
finally:
result = locals().get('result', '')
connection = httplib2.Http()
body = urllib.urlencode({
'id': 'job',
'action': 'gfeatresponse',
'job_id': job.get('job_id', -1),
'data': json.dumps(result)
})
headers = {'content-type': 'application/x-www-form-urlencoded'}
address = job['callback_url']
response, content = connection.request(address,
"POST",
body=body,
headers=headers)
def run_request(case, t):
files = gMiner.run(**t['kwargs'])
with track.load(files[0], chrmeta=yeast_chr_file) as x:
data = list(x.read('chr1'))
os.remove(files[0])
case.assertEqual(data, t['expected'])
import gMiner
files = gMiner.run(
track1 : '/scratch/genomic/tracks/refseq_ucsc.sql',
track1_name : 'hg19 refSeq genome-wide from UCSC',
track2 : '/scratch/genomic/tracks/hiv_bushman.sql',
track2_name : 'hg19 HIV integration sites from liftOver',
operation_type : 'genomic_manip',
manipulation : 'overlap_track',
output_location : '/tmp/gMiner/',
)
import gMiner
files = gMiner.run(
track1 = '/scratch/genomic/tracks/refseq_ucsc.sql',
track1_name = 'hg19 refSeq genome-wide from UCSC',
track2 = '/scratch/genomic/tracks/hiv_bushman.sql',
track2_name = 'hg19 HIV integration sites from liftOver',
operation_type = 'genomic_manip',
manipulation = 'overlap_track',
output_location = '/tmp/',
)
