What question did this study set out to answer?

This study aims to introduce and validate a new family of sum-connectivity descriptors for QSPR modelling.

Advancing QSPR with sum-connectivity descriptors: physicochemical and antibacterial modelling via molecular graph connectivity

View Full Paper

Ask AI

Bookmark

Authors

Randomized trial demonstrates enhanced modelling of antibacterial activity in E. coli, indicating improved prediction capability.

This study aims to introduce and validate a new family of sum-connectivity descriptors for QSPR modelling.
Developed sum-connectivity descriptors using a classical edge-degree kernel and tested them against a curated E. coli dataset from ChEMBL.
Conducted benchmarks comparing the performance of sum-connectivity indices with classical indices and various descriptor sets.
Employed machine learning models like ExtraTrees and RDKit Descriptors to evaluate prediction accuracy.
Sum-connectivity indices improved mean R^2 from 0.6951 to 0.7015 in QSPR benchmarking.
The All Combined descriptor set achieved properties-specific R^2 values between 0.9349 and 0.9997.
ExtraTrees + RDKit Descriptors model showed R^2 of 0.5959±0.0246 under repeated cross-validation.

Altairi et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark