Discovery of Novel Polycyclic Aromatic Inhibitors of Anaplastic Lymphoma Kinase (ALK) via Integrated QSAR Modeling, Machine Learning, and Pharmacophore-Guided Virtual Screening with AI-Assisted Mechanistic Interpretation of Structure–Activity Relationships

Abstract

TitleData and software for: Discovery of Novel Polycyclic Aromatic Inhibitors of Anaplastic Lymphoma Kinase (ALK) via Integrated QSAR Modeling, Machine Learning, and Pharmacophore-Guided Virtual Screening Description This repository holds the data and software files supporting the manuscript submitted to the Journal of Chemical Information and Modeling. The study builds a QSAR model (GFA-MLR) and machine-learning classifiers (Random Forest, XGBoost, k-NN, Naïve Bayes) on 287 literature ALK inhibitors (230 training, 57 test). It then validates pharmacophore hypotheses by ROC analysis and uses them to screen the NCI compound library. Hits were tested in an ALK enzyme assay and examined by docking and short molecular dynamics simulations. Contents: data/: the 287-compound dataset as CSV and Excel. It has compound names and sources, IC50 (nM and µM), pIC50, activity classes, training/test assignment and SMILES. models_and_workflows/: the GFA-MLR equation (Equation 2) with descriptor definitions, and the KNIME workflows and settings for the classifiers. pharmacophores/: the pharmacophore hypotheses used for virtual screening and ROC validation, and the pharmacophore fit values.