uniform import-export structure; sorted import groups; completed ever…

…y `__all__`; updated every `__init__`; added missing module docstrings

networkcommons/data/__init__.py

@@ -14,5 +14,5 @@
 #
 
 from ._builtin import _module_data, urls
-from .omics import *
-from .network import *
+from . import omics
+from . import network

networkcommons/data/network/lianaplus.py → networkcommons/data/network/_liana.py

@@ -13,6 +13,16 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+Prior knowledge network used by LIANA+.
+"""
+
+__all__ = ['get_lianaplus']
+
+import lazy_import
+
+liana = lazy_import.lazy_module('liana')
+
 
 def get_lianaplus(resource='Consensus'):
     """
@@ -27,8 +37,6 @@ def get_lianaplus(resource='Consensus'):
         pandas.DataFrame: Liana+ network with source, target, and sign columns.
     """
 
-    import liana
-
     network = liana.resource.select_resource(resource).drop_duplicates()
     network.columns = ['source', 'target']
     network['sign'] = 1

networkcommons/data/network/moon.py → networkcommons/data/network/_moon.py

@@ -13,26 +13,34 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+Prior knowledge network used by MOON.
+"""
+
+__all__ = ['build_moon_regulons']
+
+import lazy_import
 import numpy as np
 import pandas as pd
 
+dc = lazy_import.lazy_module('decoupler')
+
 from networkcommons import _utils
+from . import _omnipath
+from . import _liana
 
 
 def build_moon_regulons(include_liana=False):
 
-    import decoupler as dc
-    from . import omnipath
-
     dorothea_df = dc.get_collectri()
 
     TFs = np.unique(dorothea_df['source'])
 
-    full_pkn = omnipath.get_omnipath(genesymbols=True, directed_signed=True)
+    full_pkn = _omnipath.get_omnipath(genesymbols=True, directed_signed=True)
 
     if include_liana:
-        from . import lianaplus
-        ligrec_resource = lianaplus.get_lianaplus()
+
+        ligrec_resource = _liana.get_lianaplus()
 
         full_pkn = pd.concat([full_pkn, ligrec_resource])
         full_pkn['edgeID'] = full_pkn['source'] + '_' + full_pkn['target']

networkcommons/data/network/omnipath.py → networkcommons/data/network/_omnipath.py

@@ -13,12 +13,19 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
-import omnipath as op
+"""
+Access to the OmniPath database.
+"""
+
+__all__ = ['get_omnipath']
+
+import lazy_import
 import numpy as np
 
+op = lazy_import.lazy_module('omnipath')
+
 
-def get_omnipath(genesymbols=True,
-                 directed_signed=True):
+def get_omnipath(genesymbols: bool = True, directed_signed: bool = True):
     """
     Retrieves the Omnipath network with directed interactions
     and specific criteria.

networkcommons/data/omics/__init__.py

@@ -13,8 +13,12 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
-from .decryptm import *
-from .decryptm_ebi import get_decryptm as _get_decryptm
-from .deseq2 import *
-from .panacea import *
-from .scperturb import *
+"""
+Access to public omics datasets.
+"""
+
+from ._common import *
+from ._decryptm import *
+from ._deseq2 import *
+from ._panacea import *
+from ._scperturb import *

networkcommons/data/omics/common.py → networkcommons/data/omics/_common.py

@@ -13,27 +13,28 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+General procedures for downloading omics datasets.
+"""
+
 from __future__ import annotations
 
-from typing import Any, IO
+__all__ = ['datasets']
+
+from typing import IO
 import zipfile
 import os
-import re
-import glob
 import hashlib
 import contextlib
 import urllib.parse
 
-import shutil
 import requests
 import bs4
 import pandas as pd
 
-from .._builtin import _module_data
 from networkcommons import _conf
 from networkcommons._session import _log
-
-__all__ = ['datasets']
+from .._builtin import _module_data
 
 
 def _datasets() -> dict[str, dict]:

networkcommons/data/omics/decryptm.py → networkcommons/data/omics/_decryptm.py

@@ -13,18 +13,22 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+DecryptM proteomics and phosphoproteomics data.
+"""
+
 from __future__ import annotations
 
+__all__ = ['decryptm_datasets', 'decryptm_table', 'decryptm_experiment']
+
 import os
 import urllib.parse
 
 import pandas as pd
 
-from . import common
+from . import _common
 from networkcommons import _conf
 
-__all__ = ['decryptm_datasets', 'decryptm_table', 'decryptm_experiment']
-
 
 def decryptm_datasets(update: bool = False) -> pd.DataFrame:
     """
@@ -43,7 +47,7 @@ def decryptm_datasets(update: bool = False) -> pd.DataFrame:
 
     if update or not os.path.exists(path):
 
-        baseurl = urllib.parse.urljoin(common._baseurl(), 'decryptm')
+        baseurl = urllib.parse.urljoin(_common._baseurl(), 'decryptm')
 
         datasets = pd.DataFrame(
             [
@@ -52,9 +56,9 @@ def decryptm_datasets(update: bool = False) -> pd.DataFrame:
                     data_type,
                     fname,
                 )
-                for experiment in common._ls(baseurl)
-                for data_type in common._ls(f'{baseurl}/{experiment}')
-                for fname in common._ls(f'{baseurl}/{experiment}/{data_type}')
+                for experiment in _common._ls(baseurl)
+                for data_type in _common._ls(f'{baseurl}/{experiment}')
+                for fname in _common._ls(f'{baseurl}/{experiment}/{data_type}')
                 if fname.startswith('curve')
             ],
             columns = ['experiment', 'data_type', 'fname']
@@ -86,8 +90,8 @@ def decryptm_table(experiment: str, data_type: str, fname: str) -> pd.DataFrame:
         The table as a pandas data frame.
     """
 
-    return common._open(
-        common._commons_url('decryptm', **locals()),
+    return _common._open(
+        _common._commons_url('decryptm', **locals()),
         df = {'sep': '\t'},
     )
 

networkcommons/data/omics/decryptm_ebi.py → networkcommons/data/omics/_decryptm_ebi.py

@@ -15,16 +15,18 @@
 
 from __future__ import annotations
 
+__all__ = []
+
 import os
 import shutil
 import glob
 import urllib.parse
 
-from . import common
+from . import _common
 from networkcommons._session import _log
 
 
-def get_decryptm(path: str):
+def _get_decryptm(path: str):
     """
     Download DecryptM from EBI.
 
@@ -35,13 +37,13 @@ def get_decryptm(path: str):
 
     os.makedirs(path, exist_ok = True)
     url = 'https://ftp.pride.ebi.ac.uk/pride/data/archive/2023/03/PXD037285/'
-    files = [f for f in common._ls(url) if f.endswith('Curves.zip')]
+    files = [f for f in _common._ls(url) if f.endswith('Curves.zip')]
 
     for fname in files:
 
         zip_url = urllib.parse.urljoin(url, fname)
 
-        with common._open(zip_url) as zip_file:
+        with _common._open(zip_url) as zip_file:
 
             _log(f'Extracting zip `{zip_file.filename}` to `{path}`.')
             zip_file.extractall(path)

networkcommons/data/omics/deseq2.py → networkcommons/data/omics/_deseq2.py

@@ -13,8 +13,14 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+Differential expression analysis with DESeq2.
+"""
+
 from __future__ import annotations
 
+__all__ = ['deseq2']
+
 from typing import TYPE_CHECKING
 import multiprocessing
 import importlib
@@ -23,17 +29,16 @@
 
     import pandas as pd
 
+import lazy_import
 from pypath_common import _misc as _ppcommon
-import pydeseq2 as _deseq2
 
 for _mod in ('default_inference', 'dds', 'ds'):
-    importlib.import_module(f'pydeseq2.{_mod}')
+
+    globals()[f'_deseq2_{_mod}'] = lazy_import.lazy_module(f'pydeseq2.{_mod}')
 
 from networkcommons import _conf
 from networkcommons._session import _log
 
-__all__ = ['deseq2']
-
 
 def deseq2(
         counts: pd.DataFrame,
@@ -70,9 +75,9 @@ def deseq2(
     ref_group = ref_group.replace('_', '-')
 
     n_cpus = _conf.get('cpu_count', multiprocessing.cpu_count())
-    inference = _deseq2.default_inference.DefaultInference(n_cpus = n_cpus)
+    inference = _deseq2_default_inference.DefaultInference(n_cpus = n_cpus)
 
-    dds = _deseq2.dds.DeseqDataSet(
+    dds = _deseq2_dds.DeseqDataSet(
         counts=counts.T,
         metadata=metadata,
         design_factors=design_factors,
@@ -82,7 +87,7 @@ def deseq2(
 
     dds.deseq2()
 
-    results = _deseq2.ds.DeseqStats(
+    results = _deseq2_ds.DeseqStats(
         dds,
         contrast=['group', test_group, ref_group],
         inference=inference

networkcommons/data/omics/_moon.py

@@ -19,11 +19,11 @@
 
 from __future__ import annotations
 
-import pandas as pd
+__all__ = ['moon']
 
-from . import common
+import pandas as pd
 
-__all__ = ['moon']
+from . import _common
 
 
 def moon() -> dict[str, pd.DataFrame]:

networkcommons/data/omics/panacea.py → networkcommons/data/omics/_panacea.py

@@ -13,13 +13,18 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+RNA-Seq data from the 'Pancancer Analysis of Chemical Entity Activity'
+resource.
+"""
+
 from __future__ import annotations
 
-import pandas as pd
+__all__ = ['panacea']
 
-from . import common
+import pandas as pd
 
-__all__ = ['panacea']
+from . import _common
 
 
 def panacea() -> tuple[pd.DataFrame]:
@@ -31,8 +36,8 @@ def panacea() -> tuple[pd.DataFrame]:
     """
 
     return tuple(
-        common._open(
-            common._commons_url('panacea', table = table),
+        _common._open(
+            _common._commons_url('panacea', table = table),
             df = {'sep': '\t'},
         )
         for table in ('count', 'meta')

networkcommons/data/omics/scperturb.py → networkcommons/data/omics/_scperturb.py

@@ -13,15 +13,21 @@
 # https://www.gnu.org/licenses/gpl-3.0.txt
 #
 
+"""
+Single-cell RNA-Seq data from the 'scPerturb' resource.
+"""
+
 from __future__ import annotations
 
+__all__ = ['scperturb', 'scperturb_metadata', 'scperturb_datasets']
+
 from typing import Any
 import json
 
 import bs4
 import anndata as ad
 
-from . import common
+from . import _common
 
 _URL = 'https://zenodo.org/record/10044268'
 
@@ -50,7 +56,7 @@ def scperturb_metadata() -> dict[str, Any]:
     record is https://zenodo.org/records/10044268.
     """
 
-    soup = common._open(_URL, ftype = 'html')
+    soup = _common._open(_URL, ftype = 'html')
     data = soup.find(id = 'recordCitation').attrs['data-record']
 
     return json.loads(data)
@@ -72,6 +78,6 @@ def scperturb(dataset: str) -> ad.AnnData:
     """
 
     urls = scperturb_datasets()
-    path = common._maybe_download(urls[dataset])
+    path = _common._maybe_download(urls[dataset])
 
     return ad.read_h5ad(path)

networkcommons/eval/__init__.py

@@ -17,4 +17,4 @@
 Network metrics and analysis methods.
 """
 
-from . import metrics as metrics
+from .metrics import *