DrLungker: A Deep Ensemble Learning Framework for Predicting Anti-Lung Cancer Compound Activity and Validating Multitarget Potency through WaterMap, DFT, MD Simulations, and MM-GBSA Analysis

Supplementary Dataset for:DrLungker: A Deep Ensemble Learning Framework for Predicting Anti-Lung Cancer Compound Activity and Validating Multitarget Potency through WaterMap, DFT, MD Simulations, and MM-GBSA AnalysisPublished in: Advanced Theory and Simulations
Manuscript DOI: https://doi.org/10.1002/adts.202501550
More information: https://github.com/ShabanAhmad/DrLungkerDescriptionThis repository provides the full training datasets used to develop the DrLungker deep ensemble learning framework for predicting anti-lung cancer compound activity. The data were curated from PubChem and ChEMBL lung cancer bioassays, followed by preprocessing, structure standardization, descriptor generation, and quality filtering.The final dataset includes 26,396 unique compounds, each encoded with 5,883 molecular descriptors (AlvaDesc + QikProp), and was used to train the hybrid ensemble model integrating ResNet, Feedforward Neural Network (FNN), and RNN-LSTM architectures with Averaging, Voting, and Stacking techniques.These files are archived here due to their large size and ensure full reproducibility of all machine-learning results described in the manuscript.