Artificial Intelligence–Driven Experimental Pharmacology: From Target Identification to Lead Optimization

Experimental pharmacology generates increasingly complex datasets — from high-throughput screening (HTS) and high-content imaging to biophysical binding kinetics, electrophysiology, dose–response profiling, and in vivo pharmacology. Extracting reliable mechanistic insight from these data, and translating it into optimized lead compounds, remains a central challenge in drug discovery. Artificial intelligence (AI) and machine learning (ML) are emerging as powerful adjuncts to the experimental pharmacologist, capable of triaging assay outputs, interpreting subtle phenotypic signals, and guiding the next experiment in iterative discovery cycles. Distinct from purely computational drug design or omics-driven bioinformatics, the application of AI directly within the experimental pharmacology workflow — augmenting bench-side decision-making and improving the quality of pharmacological inference — represents a rapidly developing frontier with significant potential to accelerate target engagement studies, mechanism-of-action elucidation, and lead optimization.Despite growing interest in AI for drug discovery, much published work focuses either on computational chemistry (molecular generation, virtual screening) or computational biology (omics integration, network analysis), with comparatively less attention paid to AI methods that interface directly with experimental pharmacology assays and workflows. This Research Topic aims to address that gap by gathering work in which AI and ML are applied to the analysis, interpretation, and optimization of experimental pharmacological data, and to the design of the experiments that generate it.We are particularly interested in contributions demonstrating closed-loop experimental–computational cycles, AI-augmented analysis of biophysical and functional pharmacology data, machine learning models built on experimentally measured ADMET endpoints, and AI methods that improve target engagement studies, mechanism-of-action determination, and hit-to-lead progression. By centring on the synergy between machine intelligence and bench experimentation, this collection seeks to advance approaches that deepen mechanistic understanding of drug action and translate experimental pharmacology data into better optimized therapeutic candidates.This Research Topic welcomes original research, reviews, mini-reviews, methods papers, and perspectives addressing AI and ML approaches that integrate with experimental pharmacology. Studies must include an experimental pharmacology component; work based solely on in silico methods, generative molecular design, or omics-driven bioinformatics falls outside the scope of this collection and is addressed by complementary Research Topics in the journal. We particularly welcome contributions in the following areas:AI-assisted analysis of HTS, high-content imaging, and phenotypic screening data, including hit triage and false-positive reductionMachine learning applied to biophysical and target engagement assays (SPR, BLI, MST, thermal proteome profiling, cellular target engagement)AI for the interpretation of electrophysiology, ion channel, and receptor functional pharmacology dataML for dose–response curve analysis, allosteric pharmacology, and complex pharmacological readoutsClosed-loop active learning approaches integrating AI with iterative experimental synthesis and pharmacological testingAI-guided lead optimization using experimentally measured potency, selectivity, and ADMET endpointsApplication of AI to organoid, organ-on-chip, and in vivo pharmacology dataIntegration of machine learning with experimental structural pharmacology (cryo-EM, NMR, X-ray)AI methods for elucidating mechanisms of action from experimental pharmacological dataSubmissions must clearly demonstrate experimental pharmacology content and validation; purely computational, generative, or omics-focused contributions will be redirected to more appropriate Research Topics within the journal. In all cases, the primary contribution should advance understanding of drug–target interactions, pharmacological mechanisms, or therapeutic properties — submissions in which AI methodology is the central focus, with pharmacology serving only as a test dataset, fall outside the scope of this section. Topic editors: Jesús García Díaz, Kannan Sridharan Source: https://www.frontiersin.org/research-topics/82912/artificial-intelligence-driven-experimental-pharmacology-from-target-identification-to-lead-optimization