def __init__(self, args, unknown_args):
self.args = args
self.unknown_args = unknown_args
self.option = checkInput(args)
#
env.logger = getLogger(max(min(args.verbosity - 1, 2), 0),
fn=os.path.splitext(args.output[0])[0],
fv=2 if args.verbosity is not 0 else 0)
env.logger.debug('\n{0}\n{1}\n{0}'.format(
"=" * min(len(args.cmd), 100), args.cmd))
self.logger = env.logger.info if args.verbosity != 1 else printinfo
#
self.logger('Loading data from [{}] ...'.format(args.data))
if self.option == 1:
self.file = SFSFile(args.data)
else:
self.file = GFile(args.data)
self.groups = self.file.getnames()
self.logger('{:,d} units found'.format(len(self.groups)))
# load non-missing data
# to annotate to each variant position wether or not it is missing from assocation analysis
# name it chip_file because it mimics the behavior of exome chip design
if args.missing_unlisted:
self.chip_file = SFSFile(args.missing_unlisted)
else:
self.chip_file = None
# set limit
if self.args.limit:
self.limit = min(max(1, args.limit), len(self.groups))
self.logger('{:,d} units will be analyzed'.format(self.limit))
else:
self.limit = len(self.groups)
self.result = ResultManager(args.output,
action='w' if not args.append else 'a')
if self.args.verbosity == 1:
# widgets = [FormatLabel('scanning: unit %(value)d - '), BouncingBar(marker=RotatingMarker())]
widgets = [
FormatLabel('scanning: unit %(value)d - '),
Percentage(), ' ',
Bar('>'), ' ',
ETA()
]
self.pbar = ProgressBar(widgets=widgets,
maxval=self.limit,
term_width=get_terminal_size()[0] -
5).start()
else:
# use each group's progress bar or not progress bar at all
self.pbar = ProgressBarNull()
# this is buffer object to hold all input dict to a list
self.data_buffer = [] if self.args.replicates < 0 else None
def show_gdat():
if '-v2' in unknown_args:
try:
print(runCommand('ptdump {}'.format(fn)))
except:
raise ValueError(
'Cannot display summary information. Make sure "{}" exists and "ptdump" is installed'
.format(fn))
else:
try:
gf = GFile(fn)
names = gf.getnames()
gf.close()
except:
names = []
for idx, name in enumerate(names):
print('/%s' % name)
if idx >= 50 and '-v0' in unknown_args:
remaining = len(names) - 50
if remaining:
printinfo(
'%s more items not displayed. Use "-v1/-v2" switch to see more.'
% remaining)
break
def showFields(fn, border, unknown_args):
def set_style(pt):
if border == 'less':
pt.set_style(MSWORD_FRIENDLY)
if border == 'no':
pt.set_style(PLAIN_COLUMNS)
#
if fn.endswith('.csv'):
pt = from_csv(openFile(fn), delimiter=',')
set_style(pt)
header = [x for x in pt.field_names if not x.startswith('_')]
if len(unknown_args) == 0:
print('\n'.join(header))
return
fields = [
re.compile(item.replace('*', '(.*?)')) if '*' in item else item
for item in unknown_args
]
output = []
for item in fields:
if type(item) is str:
item = [x for x in header if x == item]
else:
item = [x for x in header if re.match(item, x)]
output.extend(item)
print pt.get_string(fields=output)
elif fn.endswith('.SEQPowerDB'):
if not os.path.isfile(fn):
raise OSError('Cannot find {}'.format(fn))
rs = ResultManager(fn)
pt = PrettyTable()
set_style(pt)
# show tables
if len(unknown_args) == 0:
pt.add_column('TABLES', rs.get_tables())
print pt
return
table = unknown_args[0]
if table not in rs.get_tables():
raise ValueError("Cannot find table '{}'".format(table))
if '--debug' in unknown_args:
debug = True
unknown_args.pop(unknown_args.index('--debug'))
else:
debug = False
if '--condition' in unknown_args:
fields = unknown_args[1:unknown_args.index('--condition')]
condition = ' '.join(
unknown_args[(unknown_args.index('--condition') + 1):])
else:
fields = unknown_args[1:]
condition = None
# show fields
header = sorted(rs.get_fields(table),
key=lambda x: x.replace("_", "|").replace(
'method', 'AAA').replace('power', 'AAB'))
if len(fields) == 0:
pt.add_column(table, header)
pt.align[table] = "l"
print pt
else:
names = [x for x in fields if x in header]
select_query = "SELECT {} from {} {}".format(
','.join(names), table, condition if condition else '')
if debug:
sys.stderr.write(select_query + '\n')
pt.field_names = names
for item in rs.cur.execute(select_query).fetchall():
pt.add_row(item)
print pt
elif fn.split('.')[-1] in ['gdat', 'h5', 'hdf5']:
def show_gdat():
if '-v2' in unknown_args:
try:
print(runCommand('ptdump {}'.format(fn)))
except:
raise ValueError(
'Cannot display summary information. Make sure "{}" exists and "ptdump" is installed'
.format(fn))
else:
try:
gf = GFile(fn)
names = gf.getnames()
gf.close()
except:
names = []
for idx, name in enumerate(names):
print('/%s' % name)
if idx >= 50 and '-v0' in unknown_args:
remaining = len(names) - 50
if remaining:
printinfo(
'%s more items not displayed. Use "-v1/-v2" switch to see more.'
% remaining)
break
#
if '--from' in unknown_args:
for item in unknown_args[(unknown_args.index('--from') + 1):]:
prefix, surfix = os.path.splitext(os.path.basename(item))
if surfix in ['.gdat', '.h5', '.hdf5']:
runCommand(
'h5copy -v -i {0} -o {1} -s "/{2}" -d "/{2}"'.format(
item, fn, re.sub(r'[^a-zA-Z0-9_]', '_', prefix)),
accepted_rc=[0, 1])
if not '-v0' in unknown_args:
printinfo('File {} processed!'.format(item))
if '--to' in unknown_args:
target = unknown_args[(unknown_args.index('--to') + 1):]
target = target[0] if target else os.path.splitext(fn)[0]
runCommand('mkdir -p {}'.format(target))
gf = GFile(fn)
names = gf.getnames()
gf.close()
for name in gf.getnames():
dat = GData(fn, name)
if not '-v0' in unknown_args:
printinfo('Saving files {}'.format(
os.path.join(target, '{}.*.txt'.format(name))))
dat.decompress()
for key in dat:
np.savetxt(os.path.join(target,
'{}.{}.txt'.format(name, key)),
dat[key],
fmt='%s',
delimiter='\t')
show_gdat()
else:
raise ValueError('Unsupported file type {}'.format(fn))
return
