def test_assert_ordination_results_equal(self):
minimal1 = OrdinationResults([1, 2])
# a minimal set of results should be equal to itself
assert_ordination_results_equal(minimal1, minimal1)
# type mismatch
with npt.assert_raises(AssertionError):
assert_ordination_results_equal(minimal1, 'foo')
# numeric values should be checked that they're almost equal
almost_minimal1 = OrdinationResults([1.0000001, 1.9999999])
assert_ordination_results_equal(minimal1, almost_minimal1)
# species_ids missing in one, present in the other
almost_minimal1.species_ids = ['abc', 'def']
with npt.assert_raises(AssertionError):
assert_ordination_results_equal(minimal1, almost_minimal1)
almost_minimal1.species_ids = None
# site_ids missing in one, present in the other
almost_minimal1.site_ids = ['abc', 'def']
with npt.assert_raises(AssertionError):
assert_ordination_results_equal(minimal1, almost_minimal1)
almost_minimal1.site_ids = None
# test each of the optional numeric attributes
for attr in ('species', 'site', 'biplot', 'site_constraints',
'proportion_explained'):
# missing optional numeric attribute in one, present in the other
setattr(almost_minimal1, attr, [[1, 2], [3, 4]])
with npt.assert_raises(AssertionError):
assert_ordination_results_equal(minimal1, almost_minimal1)
setattr(almost_minimal1, attr, None)
# optional numeric attributes present in both, but not almost equal
setattr(minimal1, attr, [[1, 2], [3, 4]])
setattr(almost_minimal1, attr, [[1, 2], [3.00002, 4]])
with npt.assert_raises(AssertionError):
assert_ordination_results_equal(minimal1, almost_minimal1)
setattr(minimal1, attr, None)
setattr(almost_minimal1, attr, None)
# optional numeric attributes present in both, and almost equal
setattr(minimal1, attr, [[1, 2], [3, 4]])
setattr(almost_minimal1, attr, [[1, 2], [3.00000002, 4]])
assert_ordination_results_equal(minimal1, almost_minimal1)
setattr(minimal1, attr, None)
setattr(almost_minimal1, attr, None)
def setUpClass(cls):
axis_labels = ['PC1', 'PC2', 'PC3']
cls.test_df1 = pd.DataFrame.from_dict(
{
's1': [0.1, 0.2, 7],
's2': [0.9, 0.2, 7],
},
orient='index',
columns=axis_labels,
)
cls.test_df1.index.name = 'Sample ID'
cls.pcoa1 = OrdinationResults(
'pcoa1',
'pcoa1',
eigvals=pd.Series(
[7, 2, 1],
index=axis_labels,
),
samples=cls.test_df1,
proportion_explained=pd.Series(
[0.7, 0.2, 0.1],
index=axis_labels,
),
)
cls.test_metadata = pd.DataFrame(
{
'age_cat': ['30s', '40s', '50s', '30s', None],
'num_cat': [7.24, 7.24, 8.25, 7.24, None],
'other': [1, 2, 3, 4, None],
},
index=pd.Series(['s1', 's2', 'c', 'd', 'e'], name='#SampleID'))
def setUp(self):
self.test_dm = DistanceMatrix(
np.array([
[0, 1, 2, 3, 4],
[1, 0, 4, 5, 6],
[2, 4, 0, 6, 7],
[3, 5, 6, 0, 8],
[4, 6, 7, 8, 0],
]),
ids=[f'S{i}' for i in range(5)],
)
n_samples = 100
np.random.seed(825)
sample_embedding = np.random.normal(size=(n_samples, 3)) + 2
sample_embedding[:, 1] *= 3
sample_embedding[:, 2] *= 6
sample_df = pd.DataFrame(
sample_embedding,
index=[f'S{i}' for i in range(n_samples)],
columns=[f'C{i}' for i in range(3)],
)
self.test_ord_results = OrdinationResults(
'foo',
'bar',
eigvals=pd.Series(np.arange(n_samples)),
samples=sample_df,
)
def test_str(self):
exp = ("Ordination results:\n"
"\tEigvals: 2\n"
"\tProportion explained: N/A\n"
"\tSpecies: 3x2\n"
"\tSite: 3x2\n"
"\tBiplot: N/A\n"
"\tSite constraints: N/A\n"
"\tSpecies IDs: 'Species1', 'Species2', 'Species3'\n"
"\tSite IDs: 'Site1', 'Site2', 'Site3'")
obs = str(self.ordination_results)
self.assertEqual(obs, exp)
# all optional attributes missing
exp = ("Ordination results:\n"
"\tEigvals: 1\n"
"\tProportion explained: N/A\n"
"\tSpecies: N/A\n"
"\tSite: N/A\n"
"\tBiplot: N/A\n"
"\tSite constraints: N/A\n"
"\tSpecies IDs: N/A\n"
"\tSite IDs: N/A")
obs = str(OrdinationResults(np.array([4.2])))
self.assertEqual(obs, exp)
def test_get_procrustes_results(self):
sample_id_map = {
'CP3A1': 'S1',
'CC1A1': 'S2',
'CC2A1': 'S3',
'CP1A1': 'S4'
}
actual = get_procrustes_results(StringIO(pcoa1_f),
StringIO(pcoa1_f),
sample_id_map=sample_id_map,
randomize=None,
max_dimensions=None)
# just some sanity checks as the individual componenets are
# already tested -- these are based on looking at the output of the
# run, and testing to ensure that it hasn't changed
eigvals = array([
8976580.24393, 6044862.67619, 4372581.39431, 3161360.10319,
2583594.45275, 2407555.39787
])
prop_expl = array([
23.1764657118, 15.6071186064, 11.2894866423, 8.16225689998,
6.67053450426, 6.21602253997
])
site = array([[
-0.199225958574, -0.250846540029, -0.119813087305, -0.155652031006,
0.18495315824, -0.160875399364
],
[
-0.238263544222, -0.37724227779, -0.169458651217,
0.0305157004776, 0.112181007345, 0.0677415967093
],
[
0.116737988534, 0.414627960015, 0.201315243115,
0.113769076804, -0.283025353088, -0.144278863311
],
[
0.320751514262, 0.213460857804, 0.0879564954067,
0.0113672537238, -0.0141088124974, 0.237412665966
]])
site_ids = ['S3', 'S2', 'S1', 'S4']
expected = OrdinationResults(eigvals=eigvals,
proportion_explained=prop_expl,
site=site,
site_ids=site_ids)
assert_almost_equal(actual[0].eigvals, expected.eigvals)
assert_almost_equal(actual[0].proportion_explained,
expected.proportion_explained)
self.assertEqual(actual[0].site_ids, expected.site_ids)
assert_almost_equal(actual[0].site, expected.site)
assert_almost_equal(actual[1].eigvals, expected.eigvals)
assert_almost_equal(actual[1].proportion_explained,
expected.proportion_explained)
assert_almost_equal(actual[1].site, expected.site)
self.assertEqual(actual[1].site_ids, expected.site_ids)
self.assertTrue(actual[2] < 6e-30)
def setUpClass(cls):
axis_labels = ['PC1', 'PC2', 'PC3']
cls.test_df1 = pd.DataFrame.from_dict(
{
's1': [0.1, 0.2, 7],
's2': [0.9, 0.2, 7],
},
orient='index',
columns=axis_labels,
)
cls.test_df1.index.name = 'Sample ID'
cls.pcoa1 = OrdinationResults(
'pcoa1',
'pcoa1',
eigvals=pd.Series(
[7, 2, 1],
index=axis_labels,
),
samples=cls.test_df1,
proportion_explained=pd.Series(
[0.7, 0.2, 0.1],
index=axis_labels,
),
)
cls.test_metadata = pd.DataFrame(
{
'age_cat': ['30s', '40s', '50s', '30s', None],
'num_cat': [7.24, 7.24, 8.25, 7.24, None],
'other': [1, 2, 3, 4, None],
},
index=pd.Series(['s1', 's2', 'c', 'd', 'e'], name='#SampleID'))
cls.resources = DictElement({
'datasets':
DictElement({
'dataset1':
DictElement({
'__metadata__':
MockMetadataElement(cls.test_metadata),
'__pcoa__':
PCOAElement({
'sample_set':
DictElement({
'beta_metric': cls.pcoa1,
}),
})
}),
'dataset2':
DictElement({
'__metadata__':
MockMetadataElement(cls.test_metadata),
}),
}),
})
cls.resources.accept(TrivialVisitor())
cls.res_patcher = patch(
'microsetta_public_api.api.emperor.get_resources')
cls.mock_resources = cls.res_patcher.start()
cls.mock_resources.return_value = cls.resources
def setUp(self):
# Define in-memory CA results to serialize and deserialize.
eigvals = np.array([0.0961330159181, 0.0409418140138])
species = np.array([[0.408869425742, 0.0695518116298],
[-0.1153860437, -0.299767683538],
[-0.309967102571, 0.187391917117]])
site = np.array([[-0.848956053187, 0.882764759014],
[-0.220458650578, -1.34482000302],
[1.66697179591, 0.470324389808]])
biplot = None
site_constraints = None
prop_explained = None
species_ids = ['Species1', 'Species2', 'Species3']
site_ids = ['Site1', 'Site2', 'Site3']
self.ordination_results = OrdinationResults(
eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=site_constraints,
proportion_explained=prop_explained,
species_ids=species_ids,
site_ids=site_ids)
# DataFrame for testing plot method. Has a categorical column with a
# mix of numbers and strings. Has a numeric column with a mix of ints,
# floats, and strings that can be converted to floats. Has a numeric
# column with missing data (np.nan).
self.df = pd.DataFrame([['foo', '42', 10], [22, 0, 8],
[22, -4.2, np.nan], ['foo', '42.19', 11]],
index=['A', 'B', 'C', 'D'],
columns=['categorical', 'numeric', 'nancolumn'])
# Minimal ordination results for easier testing of plotting method.
# Paired with df above.
eigvals = np.array([0.50, 0.25, 0.25])
site = np.array([[0.1, 0.2, 0.3], [0.2, 0.3, 0.4], [0.3, 0.4, 0.5],
[0.4, 0.5, 0.6]])
self.min_ord_results = OrdinationResults(eigvals=eigvals,
site=site,
site_ids=['A', 'B', 'C', 'D'])
def embed(
distance_matrix: DistanceMatrix,
n_neighbors: int,
min_dist: float = 1,
number_of_dimensions: int = 2,
random_state: int = 724,
) -> OrdinationResults:
n_samples = len(distance_matrix.ids)
if number_of_dimensions > n_samples:
raise ValueError(
f'number_of_dimensions ({number_of_dimensions}) must be fewer than'
f'number of samples ({n_samples}) - 2'
)
transformer = UMAP(
n_neighbors=n_neighbors,
n_components=number_of_dimensions,
min_dist=min_dist,
random_state=random_state,
metric='precomputed',
)
embedding = transformer.fit_transform(distance_matrix[:, :])
if embedding.shape[1] < 3:
difference = 3 - embedding.shape[1]
embedding = np.hstack((embedding, np.zeros((len(embedding),
difference))))
number_of_dimensions = embedding.shape[1]
embedding_df = pd.DataFrame(embedding, index=distance_matrix.ids,
columns=[f'UMAP-{i}' for i in
range(embedding.shape[1])]
)
null_eigvals = pd.Series(np.zeros(number_of_dimensions))
ord_results = OrdinationResults(
'umap',
'Uniform Manifold Approximation and Projection',
eigvals=null_eigvals,
samples=embedding_df,
proportion_explained=null_eigvals,
)
return center(ord_results)
def emperor_output(sklearn_output, full_file_list, eigenvalues, percent_variance, output_file, new_files = None):
print("Made it to Emperor Function!")
#read in sklearn output and format accordingly for emperor intake
eigvals = pd.Series(data = eigenvalues)
samples = pd.DataFrame(data = sklearn_output, index = full_file_list)
p_explained = pd.Series(data = percent_variance)
ores = OrdinationResults(long_method_name = "principal component analysis", short_method_name = "pcoa", eigvals = eigvals, samples = samples, proportion_explained = p_explained)
#this first part is for the global metadata file
global_metadata = pd.read_csv(config.PATH_TO_ORIGINAL_MAPPING_FILE, sep = "\t")
global_metadata_headers = global_metadata.columns.tolist()
global_metadata.rename(columns = {'filename': 'SampleID'}, inplace = True)
global_metadata["type"] = "Global Data"
global_metadata.set_index("SampleID", inplace = True)
common = global_metadata
#this part is for the user uploaded metadata file
if new_files != None:
metadata_uploaded = pd.DataFrame({"SampleID": new_files, "type":["Your Data"] * len(new_files)})
for item in global_metadata_headers:
metadata_uploaded[item] = ["Your Data"] * len(new_files)
metadata_uploaded.set_index("SampleID", inplace = True)
common = pd.concat([global_metadata, metadata_uploaded])
#so you need to align the metadata and the files contained within the ordination file BEFORE feeding it into the Emperor thing otherwise it doesn't like to output results
final_metadata, unused = common.align(samples, join = "right", axis = 0)
#call stuff to ouput an emperor plot
emp = Emperor(ores, final_metadata, remote = True)
# create an output directory
os.makedirs(output_file, exist_ok=True)
with open(os.path.join(output_file, 'index.html'), 'w') as f:
f.write(emp.make_emperor(standalone = True))
emp.copy_support_files(output_file)
def center(embedding: OrdinationResults) -> OrdinationResults:
short_name = embedding.short_method_name
long_name = embedding.long_method_name
n_dimensions = embedding.samples.shape[1]
transformer = PCA(n_components=n_dimensions)
new_embedding = transformer.fit_transform(embedding.samples)
embedding_df = pd.DataFrame(new_embedding,
index=embedding.samples.index,
columns=embedding.samples.columns
)
null_eigvals = pd.Series(np.zeros(n_dimensions))
ord_results = OrdinationResults(
short_name,
long_name,
eigvals=null_eigvals,
samples=embedding_df,
proportion_explained=null_eigvals,
)
return ord_results
def emperor_output(sklearn_output,
full_file_list,
eigenvalues,
percent_variance,
output_file,
new_files=[]):
eigvals = pd.Series(data=eigenvalues)
samples = pd.DataFrame(data=sklearn_output, index=full_file_list)
samples.index.rename("SampleID", inplace=True)
p_explained = pd.Series(data=percent_variance)
ores = OrdinationResults(long_method_name="principal component analysis",
short_method_name="pcoa",
eigvals=eigvals,
samples=samples,
proportion_explained=p_explained)
#read in all sample metadata
df = pd.read_table(config.PATH_TO_ORIGINAL_MAPPING_FILE)
df.rename(columns={"filename": "SampleID"}, inplace=True)
df.set_index("SampleID", inplace=True)
#handling the case in which the pca is a projection
if len(new_files) != 0:
df["Type"] = "Global"
new_meta = pd.DataFrame({"SampleID": new_files, "Type": "Your Data"})
new_meta.set_index("SampleID", inplace=True)
df = pd.concat([df, new_meta], axis=0, join="outer")
final_metadata, unused = df.align(samples, join="right", axis=0)
#call stuff to ouput an emperor plot
emp = Emperor(ores, final_metadata, remote=True)
# create an output directory
os.makedirs(output_file, exist_ok=True)
with open(os.path.join(output_file, 'index.html'), 'w') as f:
f.write(emp.make_emperor(standalone=True))
emp.copy_support_files(output_file)
def _ordination_to_ordination_results(fh):
eigvals = _parse_vector_section(fh, 'Eigvals')
if eigvals is None:
raise OrdinationFormatError("At least one eigval must be present.")
_check_empty_line(fh)
prop_expl = _parse_vector_section(fh, 'Proportion explained')
_check_length_against_eigvals(prop_expl, eigvals,
'proportion explained values')
_check_empty_line(fh)
species, species_ids = _parse_array_section(fh, 'Species')
_check_length_against_eigvals(species, eigvals, 'coordinates per species')
_check_empty_line(fh)
site, site_ids = _parse_array_section(fh, 'Site')
_check_length_against_eigvals(site, eigvals, 'coordinates per site')
_check_empty_line(fh)
# biplot does not have ids to parse (the other arrays do)
biplot, _ = _parse_array_section(fh, 'Biplot', has_ids=False)
_check_empty_line(fh)
cons, cons_ids = _parse_array_section(fh, 'Site constraints')
if cons_ids is not None and site_ids is not None:
if cons_ids != site_ids:
raise OrdinationFormatError(
"Site constraints ids and site ids must be equal: %s != %s" %
(cons_ids, site_ids))
return OrdinationResults(eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=cons,
proportion_explained=prop_expl,
species_ids=species_ids,
site_ids=site_ids)
def _ordination_to_ordination_results(fh):
eigvals = _parse_vector_section(fh, 'Eigvals')
if eigvals is None:
raise OrdinationFormatError("At least one eigval must be present.")
_check_empty_line(fh)
prop_expl = _parse_vector_section(fh, 'Proportion explained')
_check_length_against_eigvals(prop_expl, eigvals,
'proportion explained values')
_check_empty_line(fh)
species = _parse_array_section(fh, 'Species')
_check_length_against_eigvals(species, eigvals,
'coordinates per species')
_check_empty_line(fh)
site = _parse_array_section(fh, 'Site')
_check_length_against_eigvals(site, eigvals,
'coordinates per site')
_check_empty_line(fh)
# biplot does not have ids to parse (the other arrays do)
biplot = _parse_array_section(fh, 'Biplot', has_ids=False)
_check_empty_line(fh)
cons = _parse_array_section(fh, 'Site constraints')
if cons is not None and site is not None:
if not np.array_equal(cons.index, site.index):
raise OrdinationFormatError(
"Site constraints ids and site ids must be equal: %s != %s" %
(cons.index, site.index))
return OrdinationResults(
short_method_name='', long_method_name='', eigvals=eigvals,
features=species, samples=site, biplot_scores=biplot,
sample_constraints=cons, proportion_explained=prop_expl)
def setUp(self):
super().setUp()
axis_labels = ['PC1', 'PC2', 'PC3']
self.resources = ResourceManager()
self.fh1 = self.create_tempfile(suffix='.qza')
self.fh2 = self.create_tempfile(suffix='.qza')
self.pcoa_path1 = self.fh1.name
self.pcoa_path2 = self.fh2.name
self.test_df1 = pd.DataFrame.from_dict(
{
's1': [0.1, 0.2, 7],
's2': [0.9, 0.2, 7],
},
orient='index',
columns=axis_labels,
)
self.test_df1.index.name = 'Sample ID'
self.test_df2 = pd.DataFrame.from_dict(
{
's1': [0.1, 0.2, 7],
's2': [0.9, 0.2, 7],
's3': [0.2, -0.3, 0],
's4': [0.111, -4, 0.2],
},
orient='index',
columns=axis_labels,
)
self.test_df2.index.name = 'Sample ID'
self.pcoa1 = OrdinationResults(
'pcoa1',
'pcoa1',
eigvals=pd.Series(
[7, 2, 1],
index=axis_labels,
),
samples=self.test_df1,
proportion_explained=pd.Series(
[0.7, 0.2, 0.1],
index=axis_labels,
),
)
self.pcoa2 = OrdinationResults(
'pcoa2',
'pcoa2',
eigvals=pd.Series(
[6, 3, 1],
index=axis_labels,
),
samples=self.test_df2,
proportion_explained=pd.Series(
[0.6, 0.3, 0.1],
index=axis_labels,
),
)
imported_artifact = Artifact.import_data(
"PCoAResults",
self.pcoa1,
)
imported_artifact.save(self.pcoa_path1)
imported_artifact = Artifact.import_data(
"PCoAResults",
self.pcoa2,
)
imported_artifact.save(self.pcoa_path2)
def setUp(self):
eigvals = np.array([0.512367260461, 0.300719094427, 0.267912066004,
0.208988681078, 0.19169895326, 0.16054234528,
0.15017695712, 0.122457748167, 0.0])
site = np.array([[-0.212230626531, 0.216034194368, 0.03532727349,
-0.254450494129, -0.0687468542543, 0.231895596562,
0.00496549154314, -0.0026246871695,
9.73837390723e-10],
[-0.277487312135, -0.0295483215975, -0.0744173437992,
0.0957182357964, 0.204714844022, -0.0055407341857,
-0.190287966833, 0.16307126638, 9.73837390723e-10],
[0.220886492631, 0.0874848360559, -0.351990132198,
-0.00316535032886, 0.114635191853, -0.00019194106125,
0.188557853937, 0.030002427212, 9.73837390723e-10],
[0.0308923744062, -0.0446295973489, 0.133996451689,
0.29318228566, -0.167812539312, 0.130996149793,
0.113551017379, 0.109987942454, 9.73837390723e-10],
[0.27616778138, -0.0341866951102, 0.0633000238256,
0.100446653327, 0.123802521199, 0.1285839664,
-0.132852841046, -0.217514322505, 9.73837390723e-10],
[0.202458130052, -0.115216120518, 0.301820871723,
-0.18300251046, 0.136208248567, -0.0989435556722,
0.0927738484879, 0.0909429797672, 9.73837390723e-10],
[0.236467470907, 0.21863434374, -0.0301637746424,
-0.0225473129718, -0.205287183891, -0.180224615141,
-0.165277751908, 0.0411933458557, 9.73837390723e-10],
[-0.105517545144, -0.41405687433, -0.150073017617,
-0.116066751485, -0.158763393475, -0.0223918378516,
-0.0263068046112, -0.0501209518091,
9.73837390723e-10],
[-0.371636765565, 0.115484234741, 0.0721996475289,
0.0898852445906, 0.0212491652909, -0.184183028843,
0.114877153051, -0.164938000185, 9.73837390723e-10]])
prop_expl = np.array([25.6216900347, 15.7715955926, 14.1215046787,
11.6913885817, 9.83044890697, 8.51253468595,
7.88775505332, 6.56308246609, 4.42499350906e-16])
site_ids = ['PC.636', 'PC.635', 'PC.356', 'PC.481', 'PC.354', 'PC.593',
'PC.355', 'PC.607', 'PC.634']
self.ord_res = OrdinationResults(eigvals=eigvals, site=site,
proportion_explained=prop_expl,
site_ids=site_ids)
metadata_map = {'PC.354': {'Treatment': 'Control',
'DOB': '20061218',
'Weight': '60',
'Description': 'Control_mouse_I.D._354'},
'PC.355': {'Treatment': 'Control',
'DOB': '20061218',
'Weight': '55',
'Description': 'Control_mouse_I.D._355'},
'PC.356': {'Treatment': 'Control',
'DOB': '20061126',
'Weight': '50',
'Description': 'Control_mouse_I.D._356'},
'PC.481': {'Treatment': 'Control',
'DOB': '20070314',
'Weight': '52',
'Description': 'Control_mouse_I.D._481'},
'PC.593': {'Treatment': 'Control',
'DOB': '20071210',
'Weight': '57',
'Description': 'Control_mouse_I.D._593'},
'PC.607': {'Treatment': 'Fast',
'DOB': '20071112',
'Weight': '65',
'Description': 'Fasting_mouse_I.D._607'},
'PC.634': {'Treatment': 'Fast',
'DOB': '20080116',
'Weight': '68',
'Description': 'Fasting_mouse_I.D._634'},
'PC.635': {'Treatment': 'Fast',
'DOB': '20080116',
'Weight': '70',
'Description': 'Fasting_mouse_I.D._635'},
'PC.636': {'Treatment': 'Fast',
'DOB': '20080116',
'Weight': '72',
'Description': 'Fasting_mouse_I.D._636'}}
self.metadata_map = pd.DataFrame.from_dict(metadata_map,
orient='index')
self.categories = ['Treatment']
self.sort_by = 'Weight'
def setUp(self):
super(OrdinationResultsReaderWriterTests, self).setUp()
# define in-memory results, one for each of the valid files in
# self.valid_fps
# CA results
eigvals = np.array([0.0961330159181, 0.0409418140138])
species = np.array([[0.408869425742, 0.0695518116298],
[-0.1153860437, -0.299767683538],
[-0.309967102571, 0.187391917117]])
site = np.array([[-0.848956053187, 0.882764759014],
[-0.220458650578, -1.34482000302],
[1.66697179591, 0.470324389808]])
biplot = None
site_constraints = None
prop_explained = None
species_ids = ['Species1', 'Species2', 'Species3']
site_ids = ['Site1', 'Site2', 'Site3']
ca_scores = OrdinationResults(eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=site_constraints,
proportion_explained=prop_explained,
species_ids=species_ids,
site_ids=site_ids)
# CCA results
eigvals = np.array([
0.366135830393, 0.186887643052, 0.0788466514249, 0.082287840501,
0.0351348475787, 0.0233265839374, 0.0099048981912,
0.00122461669234, 0.000417454724117
])
species = np.loadtxt(get_data_path('ordres_exp_OrdRes_CCA_species'))
site = np.loadtxt(get_data_path('ordres_exp_OrdRes_CCA_site'))
biplot = np.array(
[[-0.169746767979, 0.63069090084, 0.760769036049],
[-0.994016563505, 0.0609533148724, -0.0449369418179],
[0.184352565909, -0.974867543612, 0.0309865007541]])
site_constraints = np.loadtxt(
get_data_path('ordres_exp_OrdRes_CCA_site_constraints'))
prop_explained = None
species_ids = [
'Species0', 'Species1', 'Species2', 'Species3', 'Species4',
'Species5', 'Species6', 'Species7', 'Species8'
]
site_ids = [
'Site0', 'Site1', 'Site2', 'Site3', 'Site4', 'Site5', 'Site6',
'Site7', 'Site8', 'Site9'
]
cca_scores = OrdinationResults(eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=site_constraints,
proportion_explained=prop_explained,
species_ids=species_ids,
site_ids=site_ids)
# PCoA results
eigvals = np.array([
0.512367260461, 0.300719094427, 0.267912066004, 0.208988681078,
0.19169895326, 0.16054234528, 0.15017695712, 0.122457748167, 0.0
])
species = None
site = np.loadtxt(get_data_path('ordres_exp_OrdRes_PCoA_site'))
biplot = None
site_constraints = None
prop_explained = np.array([
0.267573832777, 0.15704469605, 0.139911863774, 0.109140272454,
0.100111048503, 0.0838401161912, 0.0784269939011, 0.0639511763509,
0.0
])
species_ids = None
site_ids = [
'PC.636', 'PC.635', 'PC.356', 'PC.481', 'PC.354', 'PC.593',
'PC.355', 'PC.607', 'PC.634'
]
pcoa_scores = OrdinationResults(eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=site_constraints,
proportion_explained=prop_explained,
species_ids=species_ids,
site_ids=site_ids)
# RDA results
eigvals = np.array([
25.8979540892, 14.9825779819, 8.93784077262, 6.13995623072,
1.68070536498, 0.57735026919, 0.275983624351
])
species = np.loadtxt(get_data_path('ordres_exp_OrdRes_RDA_species'))
site = np.loadtxt(get_data_path('ordres_exp_OrdRes_RDA_site'))
biplot = np.array([[0.422650019179, -0.559142585857, -0.713250678211],
[0.988495963777, 0.150787422017, -0.0117848614073],
[-0.556516618887, 0.817599992718, 0.147714267459],
[-0.404079676685, -0.9058434809, -0.127150316558]])
site_constraints = np.loadtxt(
get_data_path('ordres_exp_OrdRes_RDA_site_constraints'))
prop_explained = None
species_ids = [
'Species0', 'Species1', 'Species2', 'Species3', 'Species4',
'Species5'
]
site_ids = [
'Site0', 'Site1', 'Site2', 'Site3', 'Site4', 'Site5', 'Site6',
'Site7', 'Site8', 'Site9'
]
rda_scores = OrdinationResults(eigvals=eigvals,
species=species,
site=site,
biplot=biplot,
site_constraints=site_constraints,
proportion_explained=prop_explained,
species_ids=species_ids,
site_ids=site_ids)
self.ordination_results_objs = [
ca_scores, cca_scores, pcoa_scores, rda_scores
]
def get_procrustes_results(coords_f1, coords_f2, sample_id_map=None,
randomize=None, max_dimensions=None,
get_eigenvalues=get_mean_eigenvalues,
get_percent_variation_explained=get_mean_percent_variation):
""" """
# Parse the PCoA files
ord_res_1 = OrdinationResults.read(coords_f1)
ord_res_2 = OrdinationResults.read(coords_f2)
sample_ids1 = ord_res_1.site_ids
coords1 = ord_res_1.site
eigvals1 = ord_res_1.eigvals
pct_var1 = ord_res_1.proportion_explained
sample_ids2 = ord_res_2.site_ids
coords2 = ord_res_2.site
eigvals2 = ord_res_2.eigvals
pct_var2 = ord_res_2.proportion_explained
if sample_id_map:
sample_ids1 = map_sample_ids(sample_ids1, sample_id_map)
sample_ids2 = map_sample_ids(sample_ids2, sample_id_map)
# rearrange the order of coords in coords2 to correspond to
# the order of coords in coords1
order = list(set(sample_ids1) & set(sample_ids2))
coords1 = reorder_coords(coords1, sample_ids1, order)
coords2 = reorder_coords(coords2, sample_ids2, order)
if len(order) == 0:
raise ValueError('No overlapping samples in the two files')
# If this is a random trial, apply the shuffling function passed as
# randomize()
if randomize:
coords2 = randomize(coords2)
randomized_coords2 = OrdinationResults(eigvals=eigvals2,
proportion_explained=pct_var2,
site=coords2,
site_ids=order)
else:
randomized_coords2 = None
coords1, coords2 = pad_coords_matrices(coords1, coords2)
if max_dimensions:
coords1 = filter_coords_matrix(coords1, max_dimensions)
coords2 = filter_coords_matrix(coords2, max_dimensions)
pct_var1 = pct_var1[:max_dimensions]
pct_var2 = pct_var2[:max_dimensions]
eigvals1 = eigvals1[:max_dimensions]
eigvals2 = eigvals2[:max_dimensions]
else:
if len(pct_var1) > len(pct_var2):
pct_var2 = append(pct_var2, zeros(len(pct_var1) - len(pct_var2)))
eigvals2 = append(eigvals2, zeros(len(eigvals1) - len(eigvals2)))
elif len(pct_var1) < len(pct_var2):
pct_var1 = append(pct_var1, zeros(len(pct_var2) - len(pct_var1)))
eigvals1 = append(eigvals1, zeros(len(eigvals2) - len(eigvals1)))
# Run the Procrustes analysis
transformed_coords_m1, transformed_coords_m2, m_squared =\
procrustes(coords1, coords2)
# print coords2
# print transformed_coords_m2
eigvals = get_eigenvalues(eigvals1, eigvals2)
pct_var = get_percent_variation_explained(pct_var1, pct_var2)
transformed_coords1 = OrdinationResults(eigvals=asarray(eigvals),
proportion_explained=asarray(pct_var),
site=asarray(transformed_coords_m1),
site_ids=order)
transformed_coords2 = OrdinationResults(eigvals=asarray(eigvals),
proportion_explained=asarray(pct_var),
site=asarray(transformed_coords_m2),
site_ids=order)
# Return the results
return (transformed_coords1, transformed_coords2,
m_squared, randomized_coords2)
