Related Experiment Videos
Similarity of molecular shape
1Physical Chemistry Laboratory, Oxford, U.K.
Journal of Computer-Aided Molecular Design
|October 1, 1991
Summary
This study introduces a novel method to quantify molecular similarity based on shape, simplifying calculations and enabling efficient analysis of isomers. The approach also provides a new way to grade chiral molecules by their dissimilarity.
Area of Science:
- Computational Chemistry
- Molecular Modeling
- Cheminformatics
Background:
- Traditional molecular similarity assessments rely on electron densities or electrostatic potentials.
- These methods can be computationally intensive and complex to implement.
- A need exists for faster, simpler, and shape-based similarity metrics.
Purpose of the Study:
- To develop a novel, shape-based quantitative measure of molecular similarity.
- To introduce 'elements of similarity' analogous to 'elements of chirality'.
- To provide a computationally efficient method for assessing molecular similarity, particularly for isomers and enantiomers.
Main Methods:
- Formulating molecular similarity (S) as a direct function of molecular shape.
- Developing methods to bypass time-consuming orientation optimization between molecules.
- Identifying 'elements of similarity' that are variable properties.
Main Results:
- The proposed shape-based similarity formulations are simple and computationally fast.
- The method effectively quantifies the similarity of isomers.
- Dissimilarity (D = 1 - S) is shown to be a shape-analog of the 'chirality coefficient' for enantiomeric pairs.
Conclusions:
- Shape-based molecular similarity offers a computationally efficient alternative to traditional methods.
- The concept of 'elements of similarity' provides a new framework for molecular comparison.
- The dissimilarity metric (D) offers a novel approach to grading chiral molecules.