An electrotopological-state index for atoms in molecules

Area of Science:

• Computational Chemistry
• Cheminformatics
• Molecular Modeling

Background:

• Molecular structure description is crucial for predicting chemical and biological properties.
• Existing methods often focus on either electronic or topological aspects, limiting comprehensive analysis.

Purpose of the Study:

• To introduce a new method for molecular structure description integrating electronic and topological features.
• To validate the E-state method's applicability in predicting spectroscopic properties and biological activities.

Main Methods:

• Utilizing a hydrogen-suppressed graph to represent molecular skeletons.
• Assigning intrinsic atom values (I) based on valence and sigma electrons.
• Calculating electrotopological state (E-state) values (Si) by incorporating inter-atomic influences.

Main Results:

• Demonstrated the E-state method's effectiveness across diverse organic structures (e.g., branching, heteroatoms).
• Achieved an excellent correlation (r=0.993) between E-state oxygen values and 17O NMR chemical shifts for alkyl ethers.
• Applied E-state values successfully in QSAR analysis for barbiturate binding to beta-cyclodextrin.

Conclusions:

• The E-state method provides a robust framework for molecular structure description by merging electronic and topological data.
• E-state values show strong predictive power for NMR chemical shifts and are valuable in QSAR studies.
• This integrated approach enhances the understanding of structure-property relationships in chemistry and biology.