liuganghuggingface/moldata-esol

ESOL / Delaney aqueous solubility

Complete untransformed DeepChem source table for ESOL / Delaney aqueous solubility. MoleculeNet is a benchmark view of the original experimental or quantum source.

Processed copy of esol from the molecular datasets collection. Structures are canonicalized and validated; labels are never averaged or filled. Prefer the original source for citation.

Where the data comes from

• Original data: published with the original study
• Original study: Delaney. ESOL: Estimating Aqueous Solubility Directly from Molecular Structure (2004)
• Redistributed by: MoleculeNet (DeepChem) (dataset page)

Data quality

Grade: care (rough; rules on the hub page).

• info: experiment. Label origin from the source's tags.
• warn: multi-source. Compiled from several labs or publications; assay protocols can differ (Walters 2023: 45% of same-compound IC50 pairs from two papers differ by >0.3 log units).
• good: valid structures. Rows RDKit cannot parse (kept in quarantine).
• warn: 19% undefined stereo. Molecules with at least one unassigned stereocenter (BACE: 71%).
• info: 2 stereoisomer sets. Groups of stereoisomers in the sampled structures; their labels can differ sharply.
• good: 1,117 labeled molecules. Median molecules with a label per task.
• good: 95% agree with aqsoldb. Aqueous solubility: on 1,099 shared molecules, 95% agree within ±0.5 log10(mol/L) (MAE 0.09, r = 0.99).
• good: 94% agree with opera_ws. Aqueous solubility: on 1,007 shared molecules, 94% agree within ±0.5 log10(mol/L) (MAE 0.11, r = 0.99).

Tables

Config	Splits	Rows	Targets
000_delaney-processed	full	1,128	measured log solubility in mols per litre (regression, log10(mol/L))

data/ holds parseable rows with identical rows collapsed. observations/ holds every source row, including unparseable structures, for auditing. Added columns: SMILES_original, SMILES_canonical (RDKit; stereo, salts, and * kept), structure_key (JSON list of canonical components; use it to deduplicate or group), structure_status, wildcard_atom_count, source_row_first, observation_count, source_dataset, source_table. All original columns are strings, so censored values such as <10 survive; cast labels yourself.

Load

from torch_molecule.datasets import load_dataset
data = load_dataset("esol")

from datasets import load_dataset
ds = load_dataset("liuganghuggingface/moldata-esol", "000_delaney-processed")

Known issues

• The ESOL predicted-solubility column is an existing model prediction, not a second experimental target. Different mirrors may contain standardized structures or different row counts.

License

Not stated by the source. Redistributed with the collection maintainer's approval; check the original source (https://github.com/deepchem/deepchem/blob/master/deepchem/molnet/load_function/delaney_datasets.py) for its terms before reuse.

Citation

Cite the original study and data source:

• Delaney. ESOL: Estimating Aqueous Solubility Directly from Molecular Structure (2004)
• Wu et al. MoleculeNet: a benchmark for molecular machine learning (2018)
• Official DeepChem loader and task definitions