Decoding MLK4 Inhibition with Interpretable Machine Learning: Identification of a Novel Dihydroimidazopyridine-Based Promising Lead Compound

An Integrated Consensus Machine Learning and Structure-Based Workflow for the Discovery of Novel Tankyrase 1 Inhibitors

The proposed workflow efficiently reduced a large chemical space to a focused set of TNKS1 inhibitor candidates while substantially reducing the experimental screening burden, highlighting the value of integrating consensus ML, SBVS, and experimental validation to accelerate early-stage hit discovery for TNKS1 and other therapeutic targets.

Discovery of a potent TDP1 inhibitor through machine learning-driven predictive modeling combined with structure-based virtual screening and experimental validation

Tyrosyl-DNA phosphodiesterase I (TDP1) repairs topoisomerase I (TOP1)–mediated DNA damage and is a promising anticancer target, particularly in combination with TOP1 inhibitors. However, the discovery of potent and drug-like TDP1 inhibitors remains challenging due to the limited structural diversity of known active compounds. Here, we developed an integrated computational framework combining machine learning (ML), deep learning (DL), and structure-based docking with experimental validation. A curated dataset of 2040 compounds (857 active, 1183 inactive) was assembled and analyzed by scaffold composition. A total of 40 binary classification models were constructed using six ML algorithms and a deep neural network (DNN), each paired with five molecular fingerprint representations, along with five graph neural network architectures (GCN, GAT, MPNN, AttentiveFP, and FPGNN). The SVM::RDKitDes model performed best (AUC = 0.89, F1 = 0.78, BA = 0.80), with robustness confirmed by Y-scrambling and randomized-split analyses, and SHAP analysis identified 20 key descriptors of TDP1 inhibition. The model was deployed as a web application (http://drugpred.top:5050) and standalone desktop applications (.exe) are available at https://github.com/zenghuang8006/TDP1-inhibitor-prediction. The validated model was applied to screen 201 231 compounds, followed by drug-likeness filtering and hierarchical docking, yielding 16 candidates. Biological evaluation identified compound AO65 as a potent TDP1 inhibitor (IC50 = 0.80 ± 0.02 µM), and quantum chemical calculations and docking elucidated its electronic properties and binding within the catalytic domain. This work demonstrates the value of integrating ML-driven prediction with structure-based approaches and identifies AO65 as a promising lead for further TDP1-focused investigation.

Identification of novel PKR inhibitory chemotypes via integrated molecular modelling and machine learning

Interferon-inducible RNA-dependent protein kinase (PKR) is an emerging therapeutic target involved in cancer, neurodegeneration, and inflammatory disorders; however, the discovery of potent and structurally diverse PKR inhibitors remains limited. In this study, we report an integrated computational–experimental strategy for the identification of novel PKR inhibitory chemotypes. Pharmacophore models were generated from flexible docking of structurally diverse reference inhibitors and validated using receiver operating characteristic (ROC) analysis. To better capture ligand flexibility and address data scarcity, multiple conformations of 223 PKR inhibitors were employed as a data augmentation strategy in machine learning-based QSAR modelling. Among several algorithms evaluated, a Naïve Bayes classifier combined with genetic function algorithm (GFA) feature selection provided the most predictive model. The optimized model, incorporating a single pharmacophore hypothesis and key physicochemical descriptors, was applied to virtual screening of the Boehringer Ingelheim opnMe compound library. Experimental validation using a PKR kinase assay identified the pre-synthesized compound BI-8128 as a potent PKR inhibitor, with an IC50 value of 174.8 nM, demonstrating higher potency than the reference inhibitor C16 under identical conditions. Notably, this compound represents a structurally distinct scaffold compared to reported PKR inhibitors, indicating effective scaffold hopping. To the best of our knowledge, this is the first report of PKR inhibitory activity for BI-8128. Overall, this work demonstrates that integrating docking-derived pharmacophores with conformational ensemble-based machine learning provides an effective approach for discovering novel inhibitors against underexplored kinase targets.

Explainable Machine Learning-Guided Virtual Screening and Molecular Docking for Identification of Novel FYN Kinase Inhibitors

FYN kinase is a non-receptor protein tyrosine kinase involved in various cancers and neurodegenerative diseases; however, no selective FYN inhibitor has been approved yet. Here we introduce the explainable Machine Learning (ML) coupled with virtual screening and Molecular Docking (MD) pipeline for fast prediction of new FYN kinase inhibitors. In this study, we constructed the training set of 906 molecules active against FYN kinase from the ChEMBL database. Molecules were encoded with Extended-Connectivity Fingerprints (ECFP4). The classification models Random Forest (RF) and eXtreme Gradient Boosting (XGBoost) were developed, and the latter showed the better performance in test (AUC=0.8118) and 5-fold cross-validation (AUC=0.8297). Based on the SHapley Additive exPlanations (SHAP) values obtained via TreeExplainer, nitrogen-containing heterocycles and hydrogen bond acceptors have been identified as the most important molecular substructures. Using the optimal XGBoost classifier, screening of 2,000 approved drugs has been performed, resulting in 470 hit molecules (23.5% hit rate). Five best molecules were further submitted to the MD procedure using AutoDock Vina to dock to FYN kinase domain (PDB RCSB: 2DQ7), showing binding energies in the interval of -9.57 to -6.32 kcal/mol. Dasatinib Anhydrous (CHEMBL1421) was the second strongest binder (-8.49 kcal/mol), effectively interacting with the ATP binding site. Although CHEMBL1171837 was the strongest binder (-9.57 kcal/mol), it was caught in the ADMET profiling. According to ADMET profiling, the top one inhibitor (CHEMBL1421) satisfies Lipinski’s rule of five and Veber rules. Analysis of hydrogen bond and hydrophobic interactions revealed hydrogen bonding with ASP148, LYS39, and ASN86 and hydrophobic interactions with ALA147, ILE80, and GLY88. Validation by self-docking procedure (self-docking or STS) showed low Root Mean Square Deviation (RMSD)<2.0 Å with a binding affinity of -11.53 kcal/mol. This work highlights how explainable ML can be used in combination with structure-based docking to expedite the drug discovery process against FYN kinase and can be applied to other kinase targets.

A synergistic deep learning and machine learning framework for screening heterocyclic compounds against ALDH1A1.

An integrated computer-aided drug design (CADD) and artificial intelligence (AI) framework to systematically identify selective ALDH1A1 inhibitors from a heterocyclic compound library is developed and suggests that LDN-27219 exhibits favorable binding characteristics and represents a promising lead candidate for subsequent experimental validation.

Targeting WEE1 kinase: an integrated machine learning–cheminformatics framework for ultra-large-scale virtual screening and novel inhibitor discovery