Commit a51911aa authored by Bharath Ramsundar's avatar Bharath Ramsundar Committed by GitHub
Browse files

Merge pull request #556 from lilleswing/perm-dataset 

Cell Membrane Permeability Dataset
parents f2adf41a a82efab0

datasets/membrane_permeability.sdf

0 → 100644
+25840 −0

File added.

Preview size limit exceeded, changes collapsed.

datasets/membrane_permeability.sdf.csv

0 → 100644
+202 −0
Original line number Diff line number Diff line 
LogP(RRCK)

-5.08

-4.82

-4.89

-4.8

-4.77

-4.85

-4.96

-5.66

-4.85

-5.8

-6

-4.48

-4.55

-5.4

-4.77

-5.4

-4.89

-4.66

-5.4

-4.62

-5.7

-4.96

-5.7

-4.66

-5.3

-4.8

-5.22

-4.68

-5.52

-6

-4.72

-4.85

-4.8

-4.57

-4.64

-5.7

-4.41

-4.92

-4.96

-5.22

-6.22

-4.66

-5.01

-5.1

-6.15

-6.3

-6.22

-5.8

-5.92

-4.82

-5.15

-5.57

-5.82

-5.85

-5.4

-5.38

-6.1

-5.01

-4.96

-5.01

-5.82

-6.05

-5.6

-5.21

-5.26

-4.71

-4.52

-4.71

-4.73

-4.99

-4.56

-4.62

-4.67

-5

-4.62

-4.6

-4.66

-4.59

-4.52

-4.84

-4.8

-4.53

-4.5

-4.79

-4.85

-6.22

-5.4

-5.34

-6.1

-6.1

-5.57

-5.64

-5.51

-5.47

-5.36

-4.74

-4.03

-4.62

-4.77

-5.32

-5.15

-5.55

-5.4

-5.41

-6.09

-5.21

-6.1

-4.56

-4.51

-4.41

-5.92

-5.54

-5.29

-5.89

-4.55

-4.99

-5.85

-5.12

-5.12

-5.15

-4.44

-4.75

-4.65

-4.59

-5.62

-4.65

-4.7

-6.1

-4.54

-6.15

-5.3

-4.8

-5.25

-4.52

-6.05

-4.74

-5.47

-4.71

-4.82

-5.38

-4.89

-4.63

-4.35

-6.3

-6.05

-4.97

-4.61

-6.05

-4.63

-4.59

-4.94

-4.57

-5.82

-4.44

-4.05

-5.34

-4.62

-4.7

-4.67

-4.55

-4.49

-5.26

-4.83

-4.96

-5.41

-6.15

-4.52

-6.05

-6.1

-4.16

-4.9

-6.22

-4.53

-4.91

-5.42

-5.39

-4.51

-5.82

-6.15

-5.21

-6.4

-5.74

-6.52

-6.7

-6.15

-5.85

-5.92

-6.22

-5.72

-6.12

-5.46

-5.31

-6.3

-5.39

-5.21

-5.8

-5.33

-5.49

-5.26

-5.89

-5.24

deepchem/data/data_loader.py

+8 −1
Original line number Diff line number Diff line
@@ -240,9 +240,16 @@ class SDFLoader(DataLoader): 
  Handles loading of SDF files.

  """



  def __init__(self, tasks, clean_mols=False, **kwargs):

    super(SDFLoader, self).__init__(tasks, **kwargs)

    self.clean_mols = clean_mols

    self.smiles_field = "smiles"

    self.mol_field = "mol"

    self.id_field = "smiles"



  def get_shards(self, input_files, shard_size):

    """Defines a generator which returns data for each shard"""

    return load_sdf_files(input_files)

    return load_sdf_files(input_files, self.clean_mols)



  def featurize_shard(self, shard):

    """Featurizes a shard of an input dataframe."""

deepchem/utils/save.py

+2 −2
Original line number Diff line number Diff line
@@ -67,7 +67,7 @@ def load_data(input_files, shard_size=None, verbose=True): 
      yield load_pickle_from_disk(input_file)





def load_sdf_files(input_files):

def load_sdf_files(input_files, clean_mols):

  """Load SDF file into dataframe."""

  dataframes = []

  for input_file in input_files:
@@ -75,7 +75,7 @@ def load_sdf_files(input_files): 
    raw_df = next(load_csv_files([input_file + ".csv"], shard_size=None))

    # Structures are stored in .sdf file

    print("Reading structures from %s." % input_file)

    suppl = Chem.SDMolSupplier(str(input_file), False, False, False)

    suppl = Chem.SDMolSupplier(str(input_file), clean_mols, False, False)

    df_rows = []

    for ind, mol in enumerate(suppl):

      if mol is not None:

examples/benchmark.py

+8 −5
Original line number Diff line number Diff line
@@ -57,6 +57,7 @@ from qm9.qm9_datasets import load_qm9 
from sampl.sampl_datasets import load_sampl

from clintox.clintox_datasets import load_clintox

from hiv.hiv_datasets import load_hiv

from membrane_permeability.membrane_permeability_datasets import load_permeability

import xgboost
@@ -92,7 +93,7 @@ def benchmark_loading_datasets(hyper_parameters, 
    mode = 'classification'

  elif dataset in [

      'kaggle', 'delaney', 'nci', 'pdbbind', 'chembl', 'qm7', 'qm7b', 'qm9',

      'sampl'

      'sampl', 'membrane_permeability'

  ]:

    mode = 'regression'

  else:
@@ -160,7 +161,8 @@ def benchmark_loading_datasets(hyper_parameters, 
      'qm9': load_qm9,

      'sampl': load_sampl,

      'clintox': load_clintox,

      'hiv': load_hiv

      'hiv': load_hiv,

      'membrane_permeability': load_permeability

  }



  print('-------------------------------------')
@@ -916,8 +918,8 @@ if __name__ == '__main__': 
      dest='dataset_args',

      default=[],

      help='Choice of dataset: tox21, sider, muv, toxcast, pcba, ' +

      'kaggle, delaney, nci, pdbbind, chembl, sampl, qm7, qm7b, qm9, clintox, hiv'

  )

      'kaggle, delaney, nci, pdbbind, chembl, sampl, qm7, qm7b, qm9, clintox, hiv,'

      ' membrane_permeability')

  parser.add_argument(

      '-t',

      action='store_true',
@@ -943,7 +945,8 @@ if __name__ == '__main__': 
  if len(datasets) == 0:

    datasets = [

        'tox21', 'sider', 'muv', 'toxcast', 'pcba', 'clintox', 'hiv', 'sampl',

        'delaney', 'nci', 'kaggle', 'pdbbind', 'chembl', 'qm7b', 'qm9'

        'delaney', 'nci', 'kaggle', 'pdbbind', 'chembl', 'qm7b', 'qm9',

        'membrane_permeability'

    ]



  #input hyperparameters

CRA Git | Maintained and supported by SUSTech CRA and CCSE