Related Experiment Videos
Rapid estimation of hydrophobicity for virtual combinatorial library analysis
1AstraZeneca R&D Mölndal, S-43183 Mölndal, Sweden. tudor.oprea@astrazeneca.com
SAR and QSAR in Environmental Research
|November 8, 2001
Summary
New NPH descriptors rapidly estimate hydrophobicity for filtering large virtual libraries. These atom-count based metrics offer efficient analysis of chemical compound libraries.
Area of Science:
- Computational Chemistry
- Medicinal Chemistry
- Drug Discovery
Background:
- Accurate hydrophobicity estimation is crucial for filtering large virtual combinatorial libraries (VCLs).
- Existing methods can be computationally intensive for extremely large datasets.
Purpose of the Study:
- To introduce novel Nonpolar, Polar, Hydrogen (NPH) descriptors for rapid hydrophobicity estimation.
- To enable efficient filtering of massive VCLs (10^6-10^12 compounds).
Main Methods:
- Development of NPH descriptors based on atom counts: P_at (polar atoms), NP_at (non-polar atoms), and SHDA (hydrogen bond donors/acceptors).
- Integration of NPH descriptors with molecular weight to define related parameters (MWP_at, MWNP_at, MWSHDA).
- Validation using Partial Least Squares (PLS) models against measured and calculated logP values.
Main Results:
- NPH descriptors demonstrated moderate-to-good predictive performance in PLS models.
- External prediction against measured logP yielded q2_pred > 0.5 (n=7954).
- External prediction against calculated logP yielded q2_pred > 0.6 (n=18991).
Conclusions:
- NPH descriptors provide a rapid and effective method for assessing hydrophobicity.
- These descriptors are suitable for preliminary analysis of extremely large VCLs.
- The NPH approach facilitates early-stage filtering before full compound enumeration.