def test_cyclone_train_on_train():
print("")
print("")
print("## Testing correctness of cyclone()")
print("")
print("# Testing algorithm on training data")
print("")
settings.enable_fastmath = True
settings.verbosity = 4
training_data = datasets.leng15(mode='sorted')
scores = pairs.cyclone(training_data,
datasets.default_cc_marker(),
iterations=1000,
min_iter=10,
min_pairs=1)
print(scores)
test_quality = utils.evaluate_prediction(prediction=scores['max_class'],
reference=ref_prediction)
print(test_quality)
assert np.allclose(np.array(test_quality.values, dtype=float),
np.ones(shape=(4, 4)),
atol=0.1)
def cyclone(
adata,
marker_pairs,
gene_names,
sample_names,
iterations=1000,
min_iter=100,
min_pairs=50):
"""Assigns scores and predicted class to observations [Scialdone15]_ [Fechtner18]_.
Calculates scores for each observation and each phase and assigns prediction
based on marker pairs indentified by sandbag.
This reproduces the approach of [Scialdone15]_ in the implementation of
[Fechtner18]_.
Parameters
----------
adata : :class:`~anndata.AnnData`
The annotated data matrix.
marker_pairs : `dict`
Dictionary of marker pairs. See :func:`~scanpy.api.sandbag` output.
gene_names: `list`
List of genes.
sample_names: `list`
List of samples.
iterations : `int`, optional (default: 1000)
An integer scalar specifying the number of
iterations for random sampling to obtain a cycle score.
min_iter : `int`, optional (default: 100)
An integer scalar specifying the minimum number of iterations
for score estimation
min_pairs : `int`, optional (default: 50)
An integer scalar specifying the minimum number of iterations
for score estimation
Returns
-------
A :class:`~pandas.DataFrame` with samples as index and categories as columns with scores for each category for each
sample and a additional column with the name of the max scoring category for each sample.
If marker pairs contain only the cell cycle categories G1, S and G2M an additional column
``pypairs_cc_prediction`` will be added. Where category S is assigned to samples where G1 and G2M score are
below 0.5.
"""
try:
from pypairs import __version__ as pypairsversion
from distutils.version import LooseVersion
if LooseVersion(pypairsversion) < LooseVersion("v3.0.9"):
raise ImportError('Please only use `pypairs` >= v3.0.9 ')
except ImportError:
raise ImportError('You need to install the package `pypairs`.')
from pypairs.pairs import cyclone
from . import settings
from pypairs import settings as pp_settings
pp_settings.verbosity = settings.verbosity
pp_settings.n_jobs = settings.n_jobs
pp_settings.writedir = settings.writedir
pp_settings.cachedir = settings.cachedir
pp_settings.logfile = settings.logfile
return cyclone(
data = adata,
marker_pairs = marker_pairs,
gene_names = gene_names,
sample_names = sample_names,
iterations = iterations,
min_iter = min_iter,
min_pairs = min_pairs
)
def cyclone(
adata: AnnData,
marker_pairs: Optional[Mapping[str, Collection[Tuple[str, str]]]] = None,
*,
iterations: int = 1000,
min_iter: int = 100,
min_pairs: int = 50,
) -> pd.DataFrame:
"""\
Assigns scores and predicted class to observations [Scialdone15]_ [Fechtner18]_.
Calculates scores for each observation and each phase and assigns prediction
based on marker pairs indentified by :func:`~scanpy.external.tl.sandbag`.
This reproduces the approach of [Scialdone15]_ in the implementation of
[Fechtner18]_.
Parameters
----------
adata
The annotated data matrix.
marker_pairs
Mapping of categories to lists of marker pairs.
See :func:`~scanpy.external.tl.sandbag` output.
iterations
An integer scalar specifying the number of
iterations for random sampling to obtain a cycle score.
min_iter
An integer scalar specifying the minimum number of iterations
for score estimation.
min_pairs
An integer scalar specifying the minimum number of pairs
for score estimation.
Returns
-------
A :class:`~pandas.DataFrame` with samples as index and categories as columns
with scores for each category for each sample and a additional column with
the name of the max scoring category for each sample.
If `marker_pairs` contains only the cell cycle categories G1, S and G2M an
additional column `pypairs_cc_prediction` will be added.
Where category S is assigned to samples where G1 and G2M score are < 0.5.
"""
_check_import()
from pypairs.pairs import cyclone
from pypairs import settings as pp_settings
pp_settings.verbosity = settings.verbosity
pp_settings.n_jobs = settings.n_jobs
pp_settings.writedir = settings.writedir
pp_settings.cachedir = settings.cachedir
pp_settings.logfile = settings.logfile
return cyclone(
data=adata,
marker_pairs=marker_pairs,
iterations=iterations,
min_iter=min_iter,
min_pairs=min_pairs,
)