def test_transform_raw_abundance(self):
"""
Testing transform_raw_abundance() function of biom_calc.py.
:return: Returns OK if testing goal is achieved, otherwise raises error.
"""
self.result = bc.transform_raw_abundance(self.biomf, sample_abd=False)
self.result1 = bc.transform_raw_abundance(self.biomf, fn=math.sqrt)
# Obtaining manual calculations for comparison testing
hand_calc = {"GG_OTU_1": 1.544068044, "GG_OTU_2": 1.579783597,
"GG_OTU_3": 1.73239376, "GG_OTU_4": 1.73239376,
"GG_OTU_5": 1.62324929}
hand_calc1 = {"S1": 5.099019514, "S2": 5.567764363, "S3": 4.242640687,
"S4": 3.31662479, "S5": 4.795831523, "S6": 5.196152423,
"S7": 5.291502622, "S8": 5.099019514, "S9": 3, "S10": 4.898979486}
# Testing the validity of transform function
for oid in hand_calc.keys():
self.assertAlmostEqual(
hand_calc[oid], self.result[oid],
msg="Raw abundance transformation not computed accurately (Test1)."
)
for sid in hand_calc1.keys():
self.assertAlmostEqual(
hand_calc1[sid], self.result1[sid],
msg="Raw abundance transformation not computed accurately (Test2)."
)
def test_transform_raw_abundance(self):
"""
Testing transform_raw_abundance() function of biom_calc.py.
:return: Returns OK if testing goal is achieved, otherwise raises
error.
"""
self.result = bc.transform_raw_abundance(
self.biom, sample_abd=False
)
self.result1 = bc.raw_abundance(self.biom, sample_abd=False)
# Obtaining manual calculations for comparison testing
hand_calc = []
for num in self.result1.values():
hand_calc.append(math.log10(float(num)))
# Testing the validity of transform function
self.assertAlmostEqual(
self.result.values(), hand_calc, places=10,
msg='Function did not calculate the transformation accurately.'
)
def test_transform_raw_abundance(self):
"""
Testing transform_raw_abundance() function of biom_calc.py.
:return: Returns OK if testing goal is achieved, otherwise raises
error.
"""
self.result = bc.transform_raw_abundance(
self.biomf, sample_abd=False
)
self.result1 = bc.raw_abundance(self.biomf, sample_abd=False)
# Obtaining manual calculations for comparison testing
hand_calc = [1.17609125906, 1.7075701761, 1.93951925262,
2.08990511144, 2.20139712432]
# Testing the validity of transform function
for hand, func in zip(hand_calc, self.result.values()):
self.assertAlmostEqual(
hand, func,
msg="Function did not calculate the transformation accurately."
)
def main():
args = handle_program_options()
try:
with open(args.otu_table):
pass
except IOError as ioe:
sys.exit(
'\nError with OTU_Sample abundance data file:{}\n'
.format(ioe)
)
try:
with open(args.mapping):
pass
except IOError as ioe:
sys.exit(
'\nError with mapping file:{}\n'
.format(ioe)
)
# input data
with open(args.otu_table) as bF:
biom = json.loads(bF.readline())
map_header, imap = util.parse_map_file(args.mapping)
# rewrite tree file with otu names
if args.input_tree:
with open(args.input_tree) as treF, open(args.output_tre, 'w') as outF:
tree = treF.readline()
if "'" in tree:
tree = tree.replace("'", '')
outF.write(newick_replace_otuids(tree, biom))
oid_rows = {row['id']: row for row in biom['rows']}
# calculate analysis results
categories = None
if args.map_categories is not None:
categories = args.map_categories.split(',')
# set transform if --stabilize_variance is specfied
tform = bc.arcsine_sqrt_transform if args.stabilize_variance else None
groups = util.gather_categories(imap, map_header, categories)
for group in groups.values():
if args.analysis_metric in ['MRA', 'NMRA']:
results = bc.MRA(biom, group.sids, transform=tform)
elif args.analysis_metric == 'raw':
results = bc.transform_raw_abundance(biom, sampleIDs=group.sids,
sample_abd=False)
group.results.update({oc.otu_name_biom(oid_rows[oid]): results[oid]
for oid in results})
# write iTol data set file
with open(args.output_itol_table, 'w') as itolF:
itolF.write('LABELS\t' + '\t'.join(groups.keys())+'\n')
itolF.write('COLORS\t{}\n'.format('\t'.join(['#ff0000'
for _ in range(len(groups))])))
all_otus = frozenset({oc.otu_name_biom(row) for row in biom['rows']})
for oname in all_otus:
row = ['{name}'] # \t{s:.2f}\t{ns:.2f}\n'
row_data = {'name': oname}
msum = 0
for name, group in groups.iteritems():
row.append('{{{}:.5f}}'.format(name))
if oname in group.results:
row_data[name] = group.results[oname]
else:
row_data[name] = 0.0
msum += row_data[name]
# normalize avg relative abundance data
if args.analysis_metric == 'NMRA' and msum > 0:
row_data.update({key: data/msum
for key, data in row_data.items()
if key != 'name'})
itolF.write('\t'.join(row).format(**row_data) + '\n')
def main():
args = handle_program_options()
try:
with open(args.otu_table):
pass
except IOError as ioe:
sys.exit(
"\nError with OTU_Sample abundance data file:{}\n".format(ioe))
try:
with open(args.mapping):
pass
except IOError as ioe:
sys.exit("\nError with mapping file:{}\n".format(ioe))
# input data
biomf = biom.load_table(args.otu_table)
map_header, imap = util.parse_map_file(args.mapping)
# rewrite tree file with otu names, skip if keep_otuids specified
if args.input_tree and not args.keep_otuids:
with open(args.input_tree) as treF, open(args.output_tre, "w") as outF:
tree = treF.readline()
if "'" in tree:
tree = tree.replace("'", '')
outF.write(newick_replace_otuids(tree, biomf))
if not args.keep_otuids:
oid_rows = {
id_: md["taxonomy"]
for val, id_, md in biomf.iter(axis="observation")
}
# calculate analysis results
categories = None
if args.map_categories is not None and args.analysis_metric != "raw":
categories = args.map_categories.split(",")
# set transform if --stabilize_variance is specfied
tform = bc.arcsine_sqrt_transform if args.stabilize_variance else None
groups = util.gather_categories(imap, map_header, categories)
for group in groups.values():
if args.analysis_metric in ["MRA", "NMRA"]:
results = bc.MRA(biomf, group.sids, transform=tform)
elif args.analysis_metric == "raw":
results = bc.transform_raw_abundance(biomf,
sampleIDs=group.sids,
sample_abd=False)
if args.keep_otuids:
group.results.update({oid: results[oid] for oid in results})
else:
group.results.update(
{oc.otu_name(oid_rows[oid]): results[oid]
for oid in results})
# write iTol data set file
with open(args.output_itol_table, "w") as itolF:
if args.analysis_metric == "raw":
itolF.write("DATASET_GRADIENT\nSEPARATOR TAB\n")
itolF.write("DATASET_LABEL\tLog Total Abundance\n")
itolF.write("COLOR\t#000000\n")
itolF.write("LEGEND_TITLE\tLog Total Abundance\n")
itolF.write("LEGEND_SHAPES\t1\n")
itolF.write("LEGEND_COLORS\t#000000\n")
itolF.write("LEGEND_LABELS\tLog Total Abundance\n")
itolF.write("COLOR_MIN\t#FFFFFF\n")
itolF.write("COLOR_MAX\t#000000\n")
else:
itolF.write("DATASET_MULTIBAR\nSEPARATOR TAB\n")
itolF.write("DATASET_LABEL\t{}\n".format(args.analysis_metric))
itolF.write("FIELD_COLORS\t{}\n".format("\t".join(
["#ff0000" for _ in range(len(groups))])))
itolF.write("FIELD_LABELS\t" + "\t".join(groups.keys()) + "\n")
itolF.write("LEGEND_TITLE\t{}\n".format(args.analysis_metric))
itolF.write("LEGEND_SHAPES\t{}\n".format("\t".join(
["1" for _ in range(len(groups))])))
itolF.write("LEGEND_COLORS\t{}\n".format("\t".join(
["#ff0000" for _ in range(len(groups))])))
itolF.write("LEGEND_LABELS\t" + "\t".join(groups.keys()) + "\n")
itolF.write("WIDTH\t300\n")
itolF.write("DATA\n")
if args.keep_otuids:
all_otus = frozenset(
{id_
for id_ in biomf.ids(axis="observation")})
else:
all_otus = frozenset({
oc.otu_name(md["taxonomy"])
for val, id_, md in biomf.iter(axis="observation")
})
for oname in all_otus:
row = ["{name}"] # \t{s:.2f}\t{ns:.2f}\n"
row_data = {"name": oname}
msum = 0
for name, group in groups.iteritems():
row.append("{{{}:.5f}}".format(name))
if oname in group.results:
row_data[name] = group.results[oname]
else:
row_data[name] = 0.0
msum += row_data[name]
# normalize avg relative abundance data
if args.analysis_metric == "NMRA" and msum > 0:
row_data.update({
key: data / msum
for key, data in row_data.items() if key != "name"
})
itolF.write("\t".join(row).format(**row_data) + "\n")
def main():
args = handle_program_options()
try:
with open(args.otu_table):
pass
except IOError as ioe:
sys.exit(
"\nError with OTU_Sample abundance data file:{}\n"
.format(ioe)
)
try:
with open(args.mapping):
pass
except IOError as ioe:
sys.exit(
"\nError with mapping file:{}\n"
.format(ioe)
)
# input data
biomf = biom.load_table(args.otu_table)
map_header, imap = util.parse_map_file(args.mapping)
# rewrite tree file with otu names
if args.input_tree:
with open(args.input_tree) as treF, open(args.output_tre, "w") as outF:
tree = treF.readline()
if "'" in tree:
tree = tree.replace("'", '')
outF.write(newick_replace_otuids(tree, biomf))
oid_rows = {id_: md["taxonomy"]
for val, id_, md in biomf.iter(axis="observation")}
# calculate analysis results
categories = None
if args.map_categories is not None:
categories = args.map_categories.split(",")
# set transform if --stabilize_variance is specfied
tform = bc.arcsine_sqrt_transform if args.stabilize_variance else None
groups = util.gather_categories(imap, map_header, categories)
for group in groups.values():
if args.analysis_metric in ["MRA", "NMRA"]:
results = bc.MRA(biomf, group.sids, transform=tform)
elif args.analysis_metric == "raw":
results = bc.transform_raw_abundance(biomf, sampleIDs=group.sids,
sample_abd=False)
group.results.update({oc.otu_name(oid_rows[oid]): results[oid]
for oid in results})
# write iTol data set file
with open(args.output_itol_table, "w") as itolF:
if args.analysis_metric == "raw":
itolF.write("DATASET_GRADIENT\nSEPARATOR TAB\n")
itolF.write("DATASET_LABEL\tLog Total Abundance\n")
itolF.write("COLOR\t#000000\n")
itolF.write("LEGEND_TITLE\tLog Total Abundance\n")
itolF.write("LEGEND_SHAPES\t1\n")
itolF.write("LEGEND_COLORS\t#000000\n")
itolF.write("LEGEND_LABELS\tLog Total Abundance\n")
itolF.write("COLOR_MIN\t#FFFFFF\n")
itolF.write("COLOR_MAX\t#000000\n")
else:
itolF.write("DATASET_MULTIBAR\nSEPARATOR TAB\n")
itolF.write("DATASET_LABEL\tNMRA\n")
itolF.write("FIELD_COLORS\t{}\n".format("\t".join(["#ff0000"
for _ in range(len(groups))])))
itolF.write("FIELD_LABELS\t" + "\t".join(groups.keys())+"\n")
itolF.write("LEGEND_TITLE\tNMRA\n")
itolF.write("LEGEND_SHAPES\t{}\n".format("\t".join(["1"
for _ in range(len(groups))])))
itolF.write("LEGEND_COLORS\t{}\n".format("\t".join(["#ff0000"
for _ in range(len(groups))])))
itolF.write("LEGEND_LABELS\t" + "\t".join(groups.keys())+"\n")
itolF.write("WIDTH\t300\n")
itolF.write("DATA\n")
all_otus = frozenset({oc.otu_name(md["taxonomy"])
for val, id_, md in
biomf.iter(axis="observation")})
for oname in all_otus:
row = ["{name}"] # \t{s:.2f}\t{ns:.2f}\n"
row_data = {"name": oname}
msum = 0
for name, group in groups.iteritems():
row.append("{{{}:.5f}}".format(name))
if oname in group.results:
row_data[name] = group.results[oname]
else:
row_data[name] = 0.0
msum += row_data[name]
# normalize avg relative abundance data
if args.analysis_metric == "NMRA" and msum > 0:
row_data.update({key: data/msum
for key, data in row_data.items()
if key != "name"})
itolF.write("\t".join(row).format(**row_data) + "\n")
