Machine learning represents a milestone in data-driven research, including material informatics, robotics, and computer-aided drug discovery. Herein, we developed variable-length-array SMILES-based (VLA-SMILES) structural descriptors for the structural description of bioactive molecules extracted from the open-access ChEMBL database. VLA-SMILES digital representation of molecular structure belongs to the class of numerically coded SMILES, while enlarging the family of binary SMILE. In this work, VLA-SMILES was found to improve the predictive ability of QSAR models based on optimized backpropagation (ATransformedBP), resilient propagation (iRPROP-), and Adam optimization learning algorithms. Robustness and the predictive ability of multilayer perceptron (MLP) models on basis of VLA-SMILES were assessed via the method of QSAR parametric model validation. In addition, the method of the statistical H0 hypothesis testing of the linear regression between real and observed activities was used for predictability estimation among VLA-SMILES featured QSAR-MLPs.

VLA-SMILES: Variable-Length-Array SMILES Descriptors in QSAR Modeling

Fall 2022 abstracts, posters, and presentations will launch in mid-September.

ACS Fall 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.