Computational Pharmacogenomics: Artificial Intelligence and Machine Learning for Variant Effect Prediction and Clinical Translation

Over the past two decades, pharmacogenomics has shifted from common, coding variants in single populations toward genome-wide, cross-population analyses that include rare and non-coding variants. This transformation has generated an unprecedented volume of complex genomic data. Computational pharmacogenomics has emerged as a critical interdisciplinary field leveraging artificial intelligence (AI), machine learning (ML), deep learning (DL), and foundation models (e.g. protein language models) to translate these massive datasets into precision medicine.This Research Topic focuses on cutting-edge machine learning and deep learning approaches designed to decode complex pharmacogenomic loci, predict drug response phenotypes, and support clinical decision-making. We particularly welcome submissions that provide open datasets, benchmark computational pipelines, or offer clinical validation of algorithmic predictions.We welcome original research, methods, reviews, and perspectives addressing:-Machine learning and deep learning architectures for predicting the functional effects of rare, non-coding, and novel genetic variants in pharmacogenes.-Algorithmic models for forecasting clinical drug response phenotypes, adverse drug events, and drug-drug-gene interactions.-Benchmarking of computational pipelines and genotyping/sequencing technologies for pharmacogenomic variant interpretation.-Clinical translation of ML-based tools, including dosing optimization, clinical decision-support systems, and real-world cohort validation.-Cross-population and multi-ancestry machine learning models to address diversity and equity in clinical implementation.Please note: If patient data are analyzed, a comprehensive description of the patients including sex, age, diagnostic criteria, inclusion and exclusion criteria, disease stage, therapy received, comorbidities as well as additional clinical information and assessment of clinical response/effects should be included. If genetic, proteomics, metabolomics, or other omics data are analyzed, a comprehensive description of the methods and the rationale for selecting the specific data studied should be provided. Studies related to natural compounds, herbal extracts, or traditional medicine products, are outside the scope of the Specialty Section Pharmacogenetics & Pharmacogenomics, and should instead be submitted to the specialty section of Ethnopharmacology. Studies solely based on the analysis of public databases or published evidence, with no further experimental insights or insufficient experimental validation, will not be included in this Research Topic.Articles should fit into the Journal’s mission & scope. Further information on the Journal’s scope can be accessed following the link https://www.frontiersin.org/journals/pharmacology/about Topic editors: Yitian Zhou, Mano Joseph Mathew, Yoomi Park Source: https://www.frontiersin.org/research-topics/84473/computational-pharmacogenomics-artificial-intelligence-and-machine-learning-for-variant-effect-prediction-and-clinical-translation