New topological indices with very high discriminatory power

Area of Science:

• * Cheminformatics
• * Computational Chemistry
• * Medicinal Chemistry

Background:

• * Quantitative structure-activity relationships (QSARs) rely on molecular descriptors.
• * The development of novel descriptors is crucial for advancing QSAR modeling.
• * Existing descriptors face challenges in uniquely identifying complex molecular structures.

Purpose of the Study:

• * To introduce two novel topological indices with enhanced discriminatory power.
• * To evaluate the uniqueness and non-degeneracy of these indices for alkane constitutional isomers.
• * To assess the intercorrelation of the new indices with established molecular descriptors.

Main Methods:

• * Calculation and application of two new topological indices.
• * Unique ranking of planar graphs for alkanes C(4) to C(6).
• * Assessment of non-degenerate values for 7668 alkane constitutional isomers (C(4)-C(15)).
• * Intercorrelation analysis with common topological indices using a diverse dataset of 820 chemicals.

Main Results:

• * The two new topological indices uniquely ranked planar graphs of alkanes C(4)-C(6).
• * Non-degenerate values were observed for all 7668 constitutional isomers from C(4) to C(15).
• * The new indices exhibited low intercorrelation with commonly used molecular descriptors, clustering separately.
• * The proposed indices demonstrated utility in QSAR modeling.

Conclusions:

• * The novel topological indices possess high discriminatory power and uniqueness.
• * These descriptors offer a valuable addition to the toolkit for QSAR modeling.
• * The low intercorrelation suggests orthogonality and complementary information content.