def action(args):
qadata = qa_from_csv(qafile)
if args.qa_file:
print(path.abspath(qafile))
elif args.variables:
_, d = qadata.popitem()
print('\n'.join(d.keys()))
elif args.names:
cmpnd_codes = dict(COMPOUND_CODES)
writer = csv.writer(sys.stdout, quoting=csv.QUOTE_NONNUMERIC)
writer.writerow(['id', 'code', 'name'])
for compound_id, d in qadata.items():
writer.writerow(
[compound_id, cmpnd_codes[compound_id], d['qa_compound']])
elif args.compound_id:
try:
for k, v in qadata[args.compound_id].items():
print('%s = %s' % (k, v))
except KeyError:
print(
'"%s" is not a valid compound id; try listing compounds using the "-n/--names" option'
% args.compound_id)
sys.exit(1)
elif args.list_calculations:
for name, description in all_checks.items():
print '%-20s %s' % (name, description)
elif args.show_calculation:
fun = getattr(calculations, args.show_calculation)
print ''.join(inspect.getsourcelines(fun)[0])
elif args.controls:
for row in CONTROL_NAMES:
print '%s\t%s' % row
elif args.algorithm:
print ''.join(inspect.getsourcelines(calculations.results)[0])
def action(args):
controls, sample_groups = get_input(args.infile, split_description=args.split_desc)
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
style = "screen" if args.outfile == sys.stdout else "file"
outfile = get_outfile(args, label="qa", ext="csv")
if args.calculate_ion_ratios:
qadata = add_ion_ratios(qadata, controls)
# if a single compound is specified, define lambda cond to use as
# a filter
if args.compound_id:
cond = lambda c: c["COMPOUND_id"] == args.compound_id
else:
cond = lambda c: True
# controls
compounds = [Compound(c, matrix, **qadata[c["COMPOUND_id"]]) for c in flatten(controls.values()) if cond(c)]
display_controls(compounds, outfile=outfile, show_all=args.show_all, message=not args.outcomes_only, style=style)
compounds = [Compound(c, matrix, **qadata[c["COMPOUND_id"]]) for c in flatten(sample_groups.values()) if cond(c)]
display_specimens(compounds, outfile=outfile, show_all=args.show_all, message=not args.outcomes_only, style=style)
if args.outfile is None:
outfile.close()
def action(args):
qadata = qa_from_csv(qafile)
if args.qa_file:
print(path.abspath(qafile))
elif args.variables:
_, d = qadata.popitem()
print('\n'.join(d.keys()))
elif args.names:
cmpnd_codes = dict(COMPOUND_CODES)
writer = csv.writer(sys.stdout, quoting=csv.QUOTE_NONNUMERIC)
writer.writerow(['id', 'code', 'name'])
for compound_id, d in qadata.items():
writer.writerow([compound_id,
cmpnd_codes[compound_id],
d['qa_compound']])
elif args.compound_id:
try:
for k,v in qadata[args.compound_id].items():
print('%s = %s' % (k,v))
except KeyError:
print('"%s" is not a valid compound id; try listing compounds using the "-n/--names" option' % args.compound_id)
sys.exit(1)
elif args.list_calculations:
for name, description in all_checks.items():
print '%-20s %s' % (name, description)
elif args.show_calculation:
fun = getattr(calculations, args.show_calculation)
print ''.join(inspect.getsourcelines(fun)[0])
elif args.controls:
for row in CONTROL_NAMES:
print '%s\t%s' % row
elif args.algorithm:
print ''.join(inspect.getsourcelines(calculations.results)[0])
def action(args):
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
# first get an ungrouped list of samples
tree = xml.etree.ElementTree.ElementTree(file=args.infile)
compound_ids = [d['qa_id'] for d in qadata.values()]
samples = [list(parse_sample(elem, compound_ids)) \
for elem in tree.getiterator('SAMPLELISTDATA')[0].findall('SAMPLE')]
compound_keys = set(['COMPOUND_id', 'COMPOUND_name',
'CONFIRMATIONIONPEAK1_area', 'ISPEAK_area', 'PEAK_analconc',
'PEAK_area', 'PEAK_foundrrt', 'PEAK_signoise', 'SAMPLE_desc',
'SAMPLE_id'])
for sample in samples:
descriptions = set(compound['SAMPLE_desc'] for compound in sample)
assert len(descriptions) == 1
desc = descriptions.pop()
# are all elements present?
for compound in sample:
if compound_keys - set(compound.keys()):
print 'sample %(SAMPLE_desc)s compound %(COMPOUND_id)s is missing' % compound,
print 'element(s)', ''.join(compound_keys - set(compound.keys()))
def test01(self):
with open('testfiles/oct24.json') as f:
controls, sample_groups = json.load(f)
qadata = qa_from_csv(qafile)
qd = add_ion_ratios(qadata, controls)
for d in qd.values():
self.assertTrue('ion_ratio_avg_calc' in d)
def test01(self):
with open('testfiles/oct24.json') as f:
controls, sample_groups = json.load(f)
qadata = qa_from_csv(qafile)
qd = add_ion_ratios(qadata, controls)
for d in qd.values():
self.assertTrue('ion_ratio_avg_calc' in d)
def action(args):
controls, sample_groups = get_input(args.infile, split_description = args.split_desc)
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
style = 'screen' if args.outfile == sys.stdout else 'file'
outfile = get_outfile(args, label = 'results', ext = 'csv')
if args.calculate_ion_ratios:
qadata = add_ion_ratios(qadata, controls)
samples = get_samples(controls, sample_groups, qadata, matrix)
display_results(samples, outfile, style = style, grouped = args.grouped)
if args.outfile is None:
outfile.close()
def action(args):
controls, sample_groups = get_input(args.infile,
split_description=args.split_desc)
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
style = 'screen' if args.outfile == sys.stdout else 'file'
outfile = get_outfile(args, label='qa', ext='csv')
if args.calculate_ion_ratios:
qadata = add_ion_ratios(qadata, controls)
# if a single compound is specified, define lambda cond to use as
# a filter
if args.compound_id:
cond = lambda c: c['COMPOUND_id'] == args.compound_id
else:
cond = lambda c: True
# controls
compounds = [
Compound(c, matrix, **qadata[c['COMPOUND_id']])
for c in flatten(controls.values()) if cond(c)
]
display_controls(compounds,
outfile=outfile,
show_all=args.show_all,
message=not args.outcomes_only,
style=style)
compounds = [
Compound(c, matrix, **qadata[c['COMPOUND_id']])
for c in flatten(sample_groups.values()) if cond(c)
]
display_specimens(compounds,
outfile=outfile,
show_all=args.show_all,
message=not args.outcomes_only,
style=style)
if args.outfile is None:
outfile.close()
def action(args):
nullchar = "." if args.outfile == sys.stdout else ""
fmt = lambda s: "%.2f" % s if isinstance(s, float) else (s or nullchar)
controls, sample_groups = get_input(args.infile)
qadata = qa_from_csv(qafile)
outfile = get_outfile(args, label="ion_ratios", ext="csv")
std_ids, std_names = zip(*[(i, n) for i, n in CONTROL_NAMES if n.startswith("std")])
headers = ["qa_id", "qa_compound"] + list(std_names) + ["ion_ratio_avg_calc", "ion_ratio_avg"]
writer = csv.DictWriter(outfile, fieldnames=headers, extrasaction="ignore")
ion_ratios = mean_ion_ratios(controls, set(std_ids))
for compound_id, d in ion_ratios.items():
if compound_id == min(ion_ratios.keys()):
writer.writerow(dict(zip(headers, headers)))
d.update(qadata[compound_id])
writer.writerow(dict((k, fmt(d.get(k))) for k in headers))
def action(args):
nullchar = '.' if args.outfile == sys.stdout else ''
fmt = lambda s: '%.2f' % s if isinstance(s, float) else (s or nullchar)
controls, sample_groups = get_input(args.infile)
qadata = qa_from_csv(qafile)
outfile = get_outfile(args, label='ion_ratios', ext='csv')
std_ids, std_names = zip(*[(i, n) for i, n in CONTROL_NAMES
if n.startswith('std')])
headers = ['qa_id', 'qa_compound'] + \
list(std_names) + ['ion_ratio_avg_calc', 'ion_ratio_avg']
writer = csv.DictWriter(outfile, fieldnames=headers, extrasaction='ignore')
ion_ratios = mean_ion_ratios(controls, set(std_ids))
for compound_id, d in ion_ratios.items():
if compound_id == min(ion_ratios.keys()):
writer.writerow(dict(zip(headers, headers)))
d.update(qadata[compound_id])
writer.writerow(dict((k, fmt(d.get(k))) for k in headers))
def action(args):
matrix = read_matrix(matrix_file)
qadata = qa_from_csv(qafile)
compound_ids, compound_codes = map(list, zip(*COMPOUND_CODES))
writer = csv.DictWriter(
args.outfile if hasattr(args.outfile, 'write') else open(args.outfile, 'w'),
fieldnames = ['infile','label'] + compound_ids,
extrasaction = 'ignore')
# create headers for the first row
d = dict(label = 'label', infile = 'infile')
d.update(dict(COMPOUND_CODES))
writer.writerow(d)
counter = count(1)
seen = defaultdict(list) # keep only the first instance of each accession
for infile in args.infiles:
d = dict(infile = infile)
controls, sample_groups = get_input(infile, split_description = args.split_desc)
if args.calculate_ion_ratios:
qadata = add_ion_ratios(qadata.copy(), controls)
patient_samples = get_samples(controls, sample_groups, qadata, matrix, quantitative = True)
for samples in patient_samples:
samples = list(samples)
# exclude any malformed samples
if any(any(c.malformed for c in sample.compounds.values()) for sample in samples):
log.warning('malformed samples in %s' % infile)
continue
# keep only the first instance of each sample
if samples[0].row_label() not in seen:
d['label'] = samples[0].row_label() if args.keep_phi else counter.next()
d.update(dict((s.COMPOUND_id, s.result) for s in samples))
writer.writerow(sanitize(d))
seen[samples[0].row_label()].append(infile)
import json
from collections import OrderedDict
from itertools import groupby
log = logging.getLogger(__name__)
from opiate.containers import Compound, Sample
from opiate.utils import flatten
from opiate.parsers import qa_from_csv, read_matrix
from opiate import qafile, matrix_file
import __init__ as config
## test data
# default qa values for this package
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
with open('testfiles/oct24.json') as f:
controls, sample_groups = json.load(f)
expt_stda = controls['stdA']
# type 'misc'
misc_sample = sample_groups['Accession01']
# type 'patient'
patient_sample = sample_groups['Accession02']
compound1 = patient_sample[0]
class TestCompound(unittest.TestCase):
import json
from itertools import chain
log = logging.getLogger(__name__)
from opiate import qafile, matrix_file
from opiate.calculations import all_checks
from opiate.parsers import qa_from_csv, read_matrix, get_samples
from opiate.display import display_specimens, display_results
from opiate.containers import Compound
from opiate.utils import flatten
import __init__ as config
from __init__ import TestCaseSuppressOutput
qadata = qa_from_csv(qafile)
matrix = read_matrix(matrix_file)
with open('testfiles/oct24.json') as f:
controls, sample_groups = json.load(f)
class TestDisplayQA(TestCaseSuppressOutput):
def test01(self):
compounds = [Compound(c, matrix, **qadata[c['COMPOUND_id']]) for c in flatten(sample_groups.values())]
display_specimens(compounds, sys.stdout)
def test02(self):
compounds = [Compound(c, matrix, **qadata[c['COMPOUND_id']]) for c in flatten(sample_groups.values())]
display_specimens(compounds, sys.stdout, message = False)
class TestDisplayResults(TestCaseSuppressOutput):
