def load_data_h5ad(self,
h5ad_file,
meta_data_file=None,
meta_data_handler=DEFAULT_METADATA,
gene_data_file=None,
gene_name_column=None,
use_layer=None):
data = anndata.read_h5ad(self.input_path(h5ad_file))
if meta_data_file is None and data.obs.shape[1] > 0:
meta_data = None
else:
meta_data = self.load_metadata_tsv(
meta_data_file,
data.obs_names,
meta_data_handler=meta_data_handler)
gene_metadata = self.load_gene_metadata_tsv(gene_data_file,
gene_name_column)
if use_layer is not None and use_layer not in data.layers:
msg = "Layer {lay} is not in {f}".format(lay=use_layer,
f=h5ad_file)
raise ValueError(msg)
# Build an InferelatorData object from a layer
elif use_layer is not None:
data = InferelatorData(data.layers[use_layer].copy(),
gene_names=data.var_names.copy(),
sample_names=data.obs_names.copy(),
meta_data=pd.concat((data.obs, meta_data),
axis=1),
gene_data=pd.concat(
(data.var, gene_metadata), axis=1))
# Build an InferelatorData object from everything
else:
data = InferelatorData(data,
meta_data=meta_data,
gene_data=gene_metadata)
# Make sure bytestrings are decoded
_safe_dataframe_decoder(data.gene_data)
_safe_dataframe_decoder(data.meta_data)
self._check_loaded_data(data, filename=h5ad_file)
return data
def load_data_mtx(self,
mtx_file,
mtx_obs=None,
mtx_feature=None,
meta_data_file=None,
meta_data_handler=DEFAULT_METADATA,
gene_data_file=None,
gene_name_column=None):
data = anndata.read_mtx(self.input_path(mtx_file))
row_names = self._load_list_from_file(
self.input_path(mtx_obs)) if mtx_obs is not None else None
col_names = self._load_list_from_file(
self.input_path(mtx_feature)) if mtx_feature is not None else None
meta_data = self.load_metadata_tsv(meta_data_file,
data.obs_names,
meta_data_handler=meta_data_handler)
gene_metadata = self.load_gene_metadata_tsv(gene_data_file,
gene_name_column)
data = InferelatorData(data,
meta_data=meta_data,
gene_data=gene_metadata,
sample_names=row_names,
gene_names=col_names)
return data
def load_data_hdf5(self,
hdf5_file,
use_layer=None,
meta_data_file=None,
meta_data_handler=DEFAULT_METADATA,
gene_data_file=None,
gene_name_column=None,
transpose_expression_data=False):
data = pd.HDFStore(self.input_path(hdf5_file), mode='r')
data = data[data.keys()[0]] if use_layer is None else data[use_layer]
meta_data = self.load_metadata_tsv(meta_data_file,
data.index,
meta_data_handler=meta_data_handler)
gene_metadata = self.load_gene_metadata_tsv(gene_data_file,
gene_name_column)
data = data.transpose() if transpose_expression_data else data
data = InferelatorData(data,
meta_data=meta_data,
gene_data=gene_metadata)
# Make sure bytestrings are decoded
_safe_dataframe_decoder(data.gene_data)
_safe_dataframe_decoder(data.meta_data)
return data
def load_data_tsv(self,
expression_matrix_file,
transpose_expression_data=False,
meta_data_file=None,
meta_data_handler=DEFAULT_METADATA,
expression_matrix_metadata=None,
gene_data_file=None,
gene_name_column=None):
Debug.vprint("Loading expression data file {file}".format(
file=expression_matrix_file),
level=0)
# Load expression data
data = self.input_dataframe(expression_matrix_file)
if expression_matrix_metadata is not None:
meta_cols = data.columns.intersection(expression_matrix_metadata)
slice_meta_data = data.loc[:, meta_cols].copy()
data = data.drop(meta_cols, axis=1)
else:
slice_meta_data = None
if meta_data_file is None and slice_meta_data is not None:
meta_data = None
else:
sample_labels = data.columns if transpose_expression_data else data.index
meta_data = self.load_metadata_tsv(
meta_data_file,
sample_labels,
meta_data_handler=meta_data_handler)
meta_data = pd.concat((meta_data, slice_meta_data), axis=1)
gene_metadata = self.load_gene_metadata_tsv(gene_data_file,
gene_name_column)
# Pack all data structures into an InferelatorData object
data = InferelatorData(data,
transpose_expression=transpose_expression_data,
meta_data=meta_data,
gene_data=gene_metadata)
self._check_loaded_data(data, filename=expression_matrix_file)
return data
from inferelator.utils.data import InferelatorData
from numpy.random import default_rng
fake_obsnames = list(map(str, range(1000)))
fake_metadata = pd.DataFrame(
{
"CONST": ["A"] * 1000,
"VAR": ["A"] * 100 + ["B"] * 200 + ["C"] * 1 + ["D"] * 99 +
["E"] * 500 + ["F"] * 100
},
index=fake_obsnames)
fake_data_object = InferelatorData(
default_rng(12345).random(size=1000).reshape((1000, 1)),
meta_data=fake_metadata,
sample_names=fake_obsnames)
TEMP_DIR = tempfile.gettempdir()
TEMP_DIR_1 = os.path.join(TEMP_DIR, "test1")
class FakeResult(object):
score = 1
name = "NAME"
all_names = ["NAME"]
all_scores = {"NAME": 1}