fix

.github/workflows/test.yml

@@ -23,7 +23,7 @@ jobs:
         run: |
           python -m pip install --upgrade pip
           pip install flake8 pytest
-          pip install --editable .
+          pip install --editable .[test]
       - name: Lint with flake8
         run: |
           # stop the build if there are Python syntax errors or undefined names

MANIFEST.in

@@ -1,5 +1,3 @@
-include deeplc/expasy/*
-include deeplc/mod_to_smiles/*
 include deeplc/mods/*
 include deeplc/package_data/**/*
 include deeplc/unimod/*

deeplc/__main__.py

@@ -12,17 +12,14 @@
 import warnings
 
 import pandas as pd
-from matplotlib import pyplot as plt
-
-from deeplc import __version__, DeepLC, FeatExtractor
-from deeplc._argument_parser import parse_arguments
-from deeplc._exceptions import DeepLCError
-
 from psm_utils.io.peptide_record import peprec_to_proforma
 from psm_utils.psm import PSM
 from psm_utils.psm_list import PSMList
 from psm_utils.io import read_file
-from psm_utils.io import write_file
+
+from deeplc import __version__, DeepLC, FeatExtractor
+from deeplc._argument_parser import parse_arguments
+from deeplc._exceptions import DeepLCError
 
 logger = logging.getLogger(__name__)
 

deeplc/deeplc.py

@@ -31,22 +31,18 @@
 
 LIBRARY = {}
 
-import os
-import sys
 import copy
 import gc
 import logging
+import math
 import multiprocessing
 import multiprocessing.dummy
-import pickle
+import random
+import sys
 import warnings
 from configparser import ConfigParser
-from tempfile import TemporaryDirectory
-from copy import deepcopy
-import random
-import math
-from collections import ChainMap
 from itertools import chain
+from tempfile import TemporaryDirectory
 
 # If CLI/GUI/frozen: disable Tensorflow info and warnings before importing
 IS_CLI_GUI = os.path.basename(sys.argv[0]) in ["deeplc", "deeplc-gui"]
@@ -65,29 +61,25 @@
 import numpy as np
 import pandas as pd
 import tensorflow as tf
-from tensorflow.python.eager import context
-from tensorflow.keras.models import load_model
-import h5py
-
-from deeplc._exceptions import CalibrationError, DeepLCError
-from deeplc.trainl3 import train_en
-
+from deeplcretrainer import deeplcretrainer
+from psm_utils.io import read_file
 from psm_utils.io.peptide_record import peprec_to_proforma
 from psm_utils.psm import PSM
 from psm_utils.psm_list import PSMList
-from psm_utils.io import read_file
-from psm_utils.io import write_file
+from tensorflow.keras.models import load_model
+from tensorflow.python.eager import context
 
-from deeplcretrainer import deeplcretrainer
+from deeplc._exceptions import CalibrationError
+from deeplc.trainl3 import train_en
 
 # "Custom" activation function
 lrelu = lambda x: tf.keras.activations.relu(x, alpha=0.1, max_value=20.0)
 
 
 try:
-    from tensorflow.compat.v1.keras.backend import set_session
+    from tensorflow.compat.v1.keras.backend import set_session  # noqa: F401
 except ImportError:
-    from tensorflow.keras.backend import set_session
+    from tensorflow.keras.backend import set_session  # noqa: F401
 try:
     from tensorflow.compat.v1.keras.backend import clear_session
 except ImportError:
@@ -112,9 +104,10 @@
 # session = tf.compat.v1.Session(config=config)
 
 # Feature extraction
-from deeplc.feat_extractor import FeatExtractor
 from pygam import LinearGAM, s
 
+from deeplc.feat_extractor import FeatExtractor
+
 
 def warn(*args, **kwargs):
     pass

deeplc/feat_extractor.py

@@ -12,35 +12,21 @@
 __email__ = ["[email protected]", "[email protected]"]
 
 # Native imports
-from operator import index
 import os
 import math
 import time
 from configparser import ConfigParser
 import ast
 from re import sub
 import logging
-from copy import deepcopy
 
-# Numpy
 import numpy as np
-
-# Pandas
 import pandas as pd
-
 from psm_utils.io.peptide_record import peprec_to_proforma
 from psm_utils.psm import PSM
 from psm_utils.psm_list import PSMList
 from pyteomics import mass
 
-from functools import lru_cache
-
-from psm_utils.io.peptide_record import peprec_to_proforma
-from psm_utils.psm import PSM
-from psm_utils.psm_list import PSMList
-from psm_utils.io import read_file
-from psm_utils.io import write_file
-
 logger = logging.getLogger(__name__)
 
 

deeplc/trainl3.py

@@ -12,35 +12,26 @@
 WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 See the License for the specific language governing permissions and
 limitations under the License.
-This code is used to train retention time predictors and store
-predictions from a CV procedure for further analysis.
 
 This project was made possible by MASSTRPLAN. MASSTRPLAN received funding 
 from the Marie Sklodowska-Curie EU Framework for Research and Innovation 
 Horizon 2020, under Grant Agreement No. 675132.
 """
 
-from sklearn.model_selection import RandomizedSearchCV
-from sklearn.linear_model import ElasticNet
-from sklearn.metrics import mean_absolute_error
-from sklearn.feature_selection import SelectFromModel
-from sklearn.model_selection import cross_val_predict
-from sklearn.model_selection import KFold
-from sklearn.base import clone
-from sklearn.model_selection import GridSearchCV
-from scipy.stats import randint
-from scipy.stats import uniform
-from numpy import arange
-from scipy.stats import pearsonr
+try:
+    from sklearn.base import clone
+    from sklearn.linear_model import ElasticNet
+    from sklearn.model_selection import GridSearchCV
+except ImportError:
+    _has_sklearn = False
+else:
+    _has_sklearn = True
 
-from operator import itemgetter
-from numpy import median
-from collections import Counter
-
-def train_en(X,y,n_jobs=16,cv=None):
+
+def train_en(X, y, n_jobs=16, cv=None):
     """
     Function that trains Layer 3 of CALLC (elastic net)
-    
+
     Parameters
     ----------
     X : pd.DataFrame
@@ -51,7 +42,7 @@ def train_en(X,y,n_jobs=16,cv=None):
         number of jobs to spawn
     cv : sklearn.model_selection.KFold
         cv object
-    
+
     Returns
     -------
     sklearn.linear_model.ElasticNet
@@ -61,35 +52,55 @@ def train_en(X,y,n_jobs=16,cv=None):
     list
         list with features used to train Layer 3
     """
-    preds = []
+    if not _has_sklearn:
+        raise ImportError(
+            "This function requires the optional dependency `scikit-learn`. Run `pip install "
+            "scikit-learn` and try again."
+        )
 
     model = ElasticNet()
     crossv_mod = clone(model)
     ret_mod = clone(model)
 
-    set_reg = [0.01,1.0,10.0,100.0,1000.0,10000.0,10000.0,100000.0,1000000.0,1000000000,1000000]
-    set_reg.extend([x/2 for x in set_reg])
-    set_reg.extend([x/3 for x in set_reg])
-
+    set_reg = [
+        0.01,
+        1.0,
+        10.0,
+        100.0,
+        1000.0,
+        10000.0,
+        10000.0,
+        100000.0,
+        1000000.0,
+        1000000000,
+        1000000,
+    ]
+    set_reg.extend([x / 2 for x in set_reg])
+    set_reg.extend([x / 3 for x in set_reg])
+
     params = {
-       'alpha': set_reg,
-       'l1_ratio' : [0.01,0.1,0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9,1.0],
-       'copy_X':[True],
-       'normalize' : [False],
-       'positive' : [True],
-       'fit_intercept'  : [True,False]
+        "alpha": set_reg,
+        "l1_ratio": [0.01, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0],
+        "copy_X": [True],
+        "normalize": [False],
+        "positive": [True],
+        "fit_intercept": [True, False],
     }
 
-    grid = GridSearchCV(model, params,cv=cv,scoring='neg_mean_absolute_error',verbose=0,n_jobs=n_jobs,refit=True)
-    grid.fit(X,y)
-
-    cv_pred = cv
-    crossv_mod.set_params(**grid.best_params_)
-    preds = cross_val_predict(crossv_mod, X=X, y=y, cv=cv_pred, n_jobs=n_jobs, verbose=0)
+    grid = GridSearchCV(
+        model,
+        params,
+        cv=cv,
+        scoring="neg_mean_absolute_error",
+        verbose=0,
+        n_jobs=n_jobs,
+        refit=True,
+    )
+    grid.fit(X, y)
 
+    crossv_mod.set_params(**grid.best_params_)
+
     ret_mod.set_params(**grid.best_params_)
-    ret_mod.fit(X,y)
+    ret_mod.fit(X, y)
 
-    coef_indexes = [i for i,coef in enumerate(ret_mod.coef_) if coef > 0.0]
-
-    return ret_mod
+    return ret_mod

pyproject.toml

@@ -0,0 +1,65 @@
+[project]
+name = "deeplc"
+version = "2.2.27"
+description = "DeepLC: Retention time prediction for (modified) peptides using Deep Learning."
+readme = "README.md"
+license = { file = "LICENSE" }
+authors = [
+    { name = "Robbin Bouwmeester", email = "[email protected]" },
+    { name = "Niels Hulstaert" },
+    { name = "Arthur Declercq" },
+    { name = "Ralf Gabriels" },
+    { name = "Lennart Martens" },
+    { name = "Sven Degroeve" },
+]
+classifiers = [
+    "Intended Audience :: Science/Research",
+    "License :: OSI Approved :: Apache Software License",
+    "Operating System :: OS Independent",
+    "Programming Language :: Python :: 3",
+    "Topic :: Scientific/Engineering :: Bio-Informatics",
+    "Development Status :: 5 - Production/Stable",
+]
+requires-python = ">=3.7"
+keywords = [
+    "DeepLC",
+    "Proteomics",
+    "deep learning",
+    "peptides",
+    "retention time",
+    "prediction",
+]
+
+dependencies = [
+    "tensorflow>=2.2,<2.13.0",
+    "numpy>=1.17,<2",
+    "pandas>=0.25,<2",
+    "h5py>=2.10.0,<4",
+    "pygam>=0.8.0,<1",
+    "deeplcretrainer>=0.1,<1",
+    "psm_utils>=0.2.3,<1",
+    "hdf5plugin>=4.1.1",
+]
+
+[project.optional-dependencies]
+test = ["pytest", "matplotlib>=3,<4"]
+gui = ["gooey>=1.0"]
+plot = ["plotly>=5"]
+deepcallc = ["scikit-learn<2,>=0.24.0"]
+
+[project.scripts]
+deeplc = "deeplc.__main__:main"
+deeplc-gui = "deeplc.gui:start_gui"
+
+[project.urls]
+GitHub = "https://github.com/compomics/deeplc"
+PyPi = "https://pypi.org/project/deeplc/"
+CompOmics = "https://www.compomics.com"
+
+[build-system]
+requires = ["setuptools"]
+build-backend = "setuptools.build_meta"
+
+[tool.setuptools]
+packages = ["deeplc"]
+include-package-data = true