Explainable QSAR Models of 5-HT1A Receptor Ligands Using Conceptual DFT Descriptors and No-Code Machine Learning Tools

This repository contains the datasets and supplementary materials supporting the study “Explainable QSAR Models of 5-HT1A Receptor Ligands Using Conceptual DFT Descriptors and No-Code Machine Learning Tools”. The dataset comprises curated molecular structures, experimental binding affinity data (Ki values and binary activity classification), and global electronic reactivity descriptors derived from Conceptual Density Functional Theory (CDFT). Descriptors were computed at the GFN1-xTB level of theory for neutral and protonated species under both vacuum and aqueous conditions. Training and external test sets are provided in ARFF format compatible with the Weka machine learning environment. These files correspond to three independent data splits used to develop and validate explainable RandomTree QSAR classification models in accordance with OECD principles for QSAR modeling. This repository also includes summary files containing model predictions and performance metrics. Together, these materials enable full reproducibility of the molecular informatics workflow and facilitate reuse of the data in ligand-based drug discovery research.