Streamlining Drug Discovery: ESOL Aqueous Solubility Prediction Now Available on Vecura
This update enables medicinal chemists and computational scientists to rapidly estimate the aqueous solubility of small molecules directly within Vecura, streamlining ADMET profiling and library triage without the need for complex local infrastructure.
What is ESOL (Delaney Aqueous Solubility)?
ESOL (Estimated SOLubility) is a classic computational model that predicts the aqueous solubility (log S) of small molecules using a straightforward multiple linear regression formula. By analyzing four key RDKit-computed descriptors—molecular weight, Crippen cLogP, number of rotatable bonds, and aromatic proportion—it estimates how easily a molecule will dissolve in water. It provides an interpretable, rapid, and CPU-efficient alternative to computationally intensive quantum-chemical calculations.
It helps users quickly estimate the solubility profile of chemical structures without needing experimental measurements. It is especially useful for early-stage drug discovery, allowing researchers to triage large virtual libraries or sanity-check synthetic targets during the design phase.
What can users do with ESOL on Vecura?
With ESOL on Vecura, users can:
- Predict Solubility Instantly: Obtain a log S value directly from a SMILES string to gauge a molecule's hydrophobicity and dissolution potential.
- Access Multiple Regression Models: Choose between the original 2004 Delaney coefficients for legacy comparison or modern "refit" coefficients for improved alignment with current RDKit standards.
- Analyze Physicochemical Descriptors: Gain transparency into the prediction by reviewing the underlying MW, cLogP, rotors, and aromatic proportion metrics for any given molecule.
- Integrate into ADMET Workflows: Feed solubility predictions into broader profiling pipelines to prioritize candidates based on their likely pharmacokinetic properties.
What the output means
The output provides a quantitative prediction (esol_log_s) of a molecule's aqueous solubility in mol/L on a logarithmic scale. A more negative value indicates lower solubility (e.g., -6 suggests the molecule is essentially insoluble), while values closer to zero or positive suggest higher solubility.
This output should be used to support scientific decision-making, such as compound prioritization and library filtering. It does not replace experimental validation, as the model is a simplified linear representation of complex solvation chemistry.
Why this matters
Aqueous solubility is a fundamental barrier in drug discovery. Compounds with poor solubility often face significant challenges in bioavailability, formulation, and clinical development. By providing a low-barrier, rapid-feedback loop on this property, the ESOL model allows medicinal chemists to address potential "solubility-limited" issues at the earliest possible stage of the design process.
While ESOL is a simplified tool with inherent limitations regarding chemical diversity and complex molecular systems, its transparency and speed make it an indispensable "first-pass" filter for virtual screening. It effectively highlights compounds that warrant further, more detailed experimental or high-level computational study.
- Developed by: Pat Walters
- Source: GitHub Repository: PatWalters/solubility
- Reference: Delaney, J. S. (2004). ESOL: Estimating Aqueous Solubility Directly from Molecular Structure. Journal of Chemical Information and Computer Sciences, 44(3), 1000–1005. https://doi.org/10.1021/ci034243x
Vecura で ESOL (Delaney Aqueous Solubility) を試す。
モデルワークスペースを開き、ご自身の入力で評価を始めましょう。