ESOL: Estimating Aqueous Solubility Directly from Molecular Structure

By J. Delaney. 2004

971 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Convolutional Networks on Graphs for Learning Molecular Fingerprints (2015, methodology): “The fingerprint network was trained as inputs to a linear model predicting solubility, as measured in [4].”
Convolutional Embedding of Attributed Molecular Graphs for Physical Property Prediction (2017, background): “solubilities in log10(mol/L).(11) The dataset of 2874 molecules was split in the original paper to three sets of “small” ( 1144), “medium” ( 485), and “large” ( 1245).”
Do Large Scale Molecular Language Representations Capture Important Structural Information? (2021, background): “ESOL Delaney (2004) is comprised of logarithmic aqueous solubility (mols/L) for 1128 organic small molecules. FreeSolv Mobley and Guthrie (2014) contains the calculated and experimental hydration free energies (kcal/mol) for 642 small neutral molecules in water.”
Unsupervised learning of dynamical and molecular similarity using variance minimization (2017, methodology): “We thus select the 59 compounds with molecular weight less than 75 g/mol from a dataset containing aqueous solubilities [Delaney, 2004].”
Pre-training Molecular Graph Representation with 3D Geometry (2021, neutral): “Molecular Property: Physical Chemistry Dataset proposed in [15] measures aqueous solubility of the molecular compounds.”