AI-Driven Modeling and Integration of Spatial and Single-Cell Omics Data

The rapid advancement of spatial and single-cell omics technologies has revolutionized molecular biosciences by enabling high-resolution profiling of gene expression and protein abundance within their native tissue microenvironments. Despite their transformative potential, these datasets are intrinsically heterogeneous, which presents significant challenges for data integration and biological interpretation. This Research Topic will explore computational modelling strategies, from statistical harmonization to deep learning, aimed at unifying multimodal omics data. By bridging experimental profiling with data-driven approaches, we seek to generate reproducible molecular maps that reveal tissue architecture, cell-type composition, and disease-associated alterations, ultimately advancing our understanding of complex biological systems.This Research Topic seeks to advance integrative modeling strategies for harmonizing spatial and single-cell omics data across platforms and conditions. Despite recent progress, the field lacks standardized approaches for normalization, batch correction, and cross-modality integration—critical steps for building reliable reference atlases. We invite contributions that develop, benchmark, or apply computational methods, including machine learning, neural networks, and statistical frameworks, to model spatially resolved molecular data. This includes applications in neuroscience, such as mapping cellular heterogeneity in the brain, resolving neuronal and glial subtypes, reconstructing spatial brain atlases, and investigating mechanisms of neurodevelopment and neurodegeneration. Emphasis will be placed on approaches that enhance reproducibility, improve biological inference, and support the construction of robust molecular maps. By fostering interdisciplinary dialogue, this collection aims to accelerate the development of scalable, generalizable tools for spatial omics analysis.We invite contributions focusing on computational, statistical, and AI-based methods for the integration, normalization, and interpretation of spatial and single-cell omics data. Topics of interest include:• Development and benchmarking of integration algorithms for spatial and single-cell datasets• Machine learning and deep learning models for spatial pattern discovery• Multi-omic data fusion and representation learning• Reference atlas construction and cross-platform harmonization• Applications combining experimental and computational modeling in tissue biology and disease• Applications to the study of neural circuit organization, cell–cell communication in the brain microenvironment, and molecular changes underlying cognitive decline and brain disordersManuscript types may include original research, methods papers, reviews, data reports, and perspectives that advance computational modeling in molecular biosciences. Interdisciplinary works bridging experimental and computational approaches are particularly encouraged. Topic editors: Martina Tarozzi, Stefano Polizzi Source: https://www.frontiersin.org/research-topics/78816/ai-driven-modeling-and-integration-of-spatial-and-single-cell-omics-data