Integrative Computational and AI Approaches for Therapeutic Discovery and Development

Computational modeling and simulation have become indispensable tools in modern therapeutic discovery and safety evaluation. The immense chemical space, coupled with the multilayered complexity of biological systems, demands integrative approaches that can predict and interpret molecular interactions across scales — from small-molecule binding and protein conformational changes to cellular and organism-level effects. Recent advances in computational chemistry, artificial intelligence (AI), and multiscale modeling have empowered researchers to explore biomolecular mechanisms in unprecedented detail. Approaches such as molecular dynamics simulations, quantum mechanical calculations, hybrid QM/MM methods, and data-driven learning models have enabled a deeper understanding of ligand–target recognition, binding energetics, and biological responses. However, despite substantial progress, gaps persist in translating computational predictions into reliable, experimentally validated therapeutic and safety outcomes. Addressing these limitations requires refining algorithms, improving interpretability, enhancing transferability across datasets and systems, and establishing robust validation frameworks that bridge computational insight with biological relevance.This Research Topic aims to gather studies that advance theoretical, computational, and AI-driven methods for biological modeling and therapeutic discovery, including the safety evaluation steps embedded within the drug development pipeline. The goal is to showcase how emerging hybrid frameworks — combining physics-based simulations and machine learning — can strengthen the prediction and optimization of biomolecular interactions, elucidate mechanisms of drug action and toxicity, and support rational therapeutic design. The Topic further encourages contributions that integrate mechanistic understanding with predictive modeling of bioactivity, metabolism, ADMET properties, or organ-specific safety profiles. Central to this objective is the development of computational strategies that enhance translational reliability, accelerate lead optimization, and reduce late-stage attrition in drug development pipelines.To gather further insights into AI-driven therapeutic modeling and safety prediction, the Topic welcomes contributions that span both methodological and application-oriented work. Studies may address, but are not limited to, the following themes:• Molecular modeling and simulation: ligand-based and structure-based modeling, molecular docking, free-energy calculations, and enhanced sampling methods.• Protein–ligand and biomolecular interaction analysis: binding mechanisms, conformational dynamics, and selectivity determinants.• Hybrid quantum mechanical and machine learning methods: QM/MM simulations, reactive modeling, machine learning–derived reactive potentials (e.g., neural network potentials trained on electronic-structure data), and electronic-structure-informed learning approaches.• AI and machine learning approaches: QSAR, graph neural networks, ensemble learning, generative models, and multi-endpoint prediction for bioactivity and toxicity.• Workflow and platform development: computational tools, databases, and automation pipelines for reproducible large-scale screening and safety assessment.• Model validation and robustness: uncertainty estimation, benchmarking, interpretability, and contextual translation to biological outcomes.Article types: Original Research, Review, Mini Review, Methods, Perspective, and Hypothesis and Theory.Topic Editor Rajiv Gandhi Govindaraj is employed by Atavistik Bio and holds patents and has published research papers related to the subject of this Research Topic. The other Topic Editors declare no potential conflicts of interest with regard to the subject of this Research Topic. Topic editors: Rajiv Gandhi Govindaraj, Prasannavenkatesh Durai, Charuvaka Muvva Source: https://www.frontiersin.org/research-topics/82128/integrative-computational-and-ai-approaches-for-therapeutic-discovery-and-development