added interaction dataset docstrings

openqdc/datasets/interaction/L7.py

@@ -10,6 +10,18 @@
 
 
 class L7(BaseInteractionDataset):
+    """
+    The L7 interaction energy dataset as described in:
+
+    Accuracy of Quantum Chemical Methods for Large Noncovalent Complexes
+    Robert Sedlak, Tomasz Janowski, Michal Pitoňák, Jan Řezáč, Peter Pulay, and Pavel Hobza
+    Journal of Chemical Theory and Computation 2013 9 (8), 3364-3374
+    DOI: 10.1021/ct400036b
+
+    Data was downloaded and extracted from:
+    http://cuby4.molecular.cz/dataset_l7.html
+    """
+
     __name__ = "L7"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"

openqdc/datasets/interaction/X40.py

@@ -10,6 +10,19 @@
 
 
 class X40(BaseInteractionDataset):
+    """
+    X40 interaction dataset of 40 dimer pairs as
+    introduced in the following paper:
+
+    Benchmark Calculations of Noncovalent Interactions of Halogenated Molecules
+    Jan Řezáč, Kevin E. Riley, and Pavel Hobza
+    Journal of Chemical Theory and Computation 2012 8 (11), 4285-4292
+    DOI: 10.1021/ct300647k
+
+    Dataset retrieved and processed from:
+    http://cuby4.molecular.cz/dataset_x40.html
+    """
+
     __name__ = "X40"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"

openqdc/datasets/interaction/des370k.py

@@ -11,6 +11,16 @@
 
 
 class DES370K(BaseInteractionDataset):
+    """
+    DE Shaw Research interaction energy of over 370K
+    small molecule dimers as described in the paper:
+
+    Quantum chemical benchmark databases of gold-standard dimer interaction energies.
+    Donchev, A.G., Taube, A.G., Decolvenaere, E. et al.
+    Sci Data 8, 55 (2021).
+    https://doi.org/10.1038/s41597-021-00833-x
+    """
+
     __name__ = "des370k_interaction"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"

openqdc/datasets/interaction/des5m.py

@@ -11,6 +11,16 @@
 
 
 class DES5M(DES370K):
+    """
+    DE Shaw Research interaction energy calculations for
+    over 5M small molecule dimers as described in the paper:
+
+    Quantum chemical benchmark databases of gold-standard dimer interaction energies.
+    Donchev, A.G., Taube, A.G., Decolvenaere, E. et al.
+    Sci Data 8, 55 (2021).
+    https://doi.org/10.1038/s41597-021-00833-x
+    """
+
     __name__ = "des5m_interaction"
     __energy_methods__ = [
         "mp2/cc-pvqz",

openqdc/datasets/interaction/dess66.py

@@ -11,6 +11,21 @@
 
 
 class DESS66(BaseInteractionDataset):
+    """
+    DE Shaw Research interaction energy
+    estimates of all 66 conformers from
+    the original S66 dataset as described
+    in the paper:
+
+    Quantum chemical benchmark databases of gold-standard dimer interaction energies.
+    Donchev, A.G., Taube, A.G., Decolvenaere, E. et al.
+    Sci Data 8, 55 (2021).
+    https://doi.org/10.1038/s41597-021-00833-x
+
+    Data was downloaded from Zenodo:
+    https://zenodo.org/records/5676284
+    """
+
     __name__ = "des_s66"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"

openqdc/datasets/interaction/dess66x8.py

@@ -11,6 +11,22 @@
 
 
 class DESS66x8(BaseInteractionDataset):
+    """
+    DE Shaw Research interaction energy
+    estimates of all 528 conformers from
+    the original S66x8 dataset as described
+    in the paper:
+
+    Quantum chemical benchmark databases of gold-standard dimer interaction energies.
+    Donchev, A.G., Taube, A.G., Decolvenaere, E. et al.
+    Sci Data 8, 55 (2021).
+    https://doi.org/10.1038/s41597-021-00833-x
+
+    Data was downloaded from Zenodo:
+
+    https://zenodo.org/records/5676284
+    """
+
     __name__ = "des_s66x8"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"

openqdc/datasets/interaction/metcalf.py

@@ -8,6 +8,23 @@
 
 
 class Metcalf(BaseInteractionDataset):
+    """
+    Hydrogen-bonded dimers of NMA with 126 molecules as described in:
+
+    Approaches for machine learning intermolecular interaction energies and
+    application to energy components from symmetry adapted perturbation theory.
+    Derek P. Metcalf, Alexios Koutsoukas, Steven A. Spronk, Brian L. Claus,
+    Deborah A. Loughney, Stephen R. Johnson, Daniel L. Cheney, C. David Sherrill;
+    J. Chem. Phys. 21 February 2020; 152 (7): 074103.
+    https://doi.org/10.1063/1.5142636
+
+    Further details:
+    "Hydrogen-bonded dimers involving N-methylacetamide (NMA) and 126 molecules
+    (46 donors and 80 acceptors; Figs. 2 and 3) were used. Optimized geometries
+    for the 126 individual monomers were obtained and paired with NMA in broad
+    arrays of spatial configurations to generate thousands of complexes for training.
+    """
+
     __name__ = "metcalf"
     __energy_unit__ = "hartree"
     __distance_unit__ = "ang"
@@ -22,14 +39,6 @@ class Metcalf(BaseInteractionDataset):
     ]
 
     def read_raw_entries(self) -> List[Dict]:
-        """
-        SMILES strings are inferred from the
-        .xyz files using the RDKit xyz2mol function.
-        More details here:
-
-        https://github.com/jensengroup/xyz2mol
-
-        """
         data = []
         for dirname in os.listdir(self.root):
             xyz_dir = os.path.join(self.root, dirname)

openqdc/datasets/interaction/splinter.py

@@ -10,6 +10,16 @@
 
 
 class Splinter(BaseInteractionDataset):
+    """
+    A dataset of over 1.7 million protein-ligand
+    interactions as described in the paper:
+
+    A quantum chemical interaction energy dataset for accurately modeling protein-ligand interactions.
+    Spronk, S.A., Glick, Z.L., Metcalf, D.P. et al.
+    Sci Data 10, 619 (2023).
+    https://doi.org/10.1038/s41597-023-02443-1
+    """
+
     __name__ = "splinter"
     __energy_methods__ = [
         "sapt0/jun-cc-pV(D+d)Z_unscaled",