We are in an era of atlas-building as we realize the incredible complexity of organisms. There has been a notable surge in efforts within academia and industry to explore the biological systems using cutting-edge technologies (e.g. MERFISH, Visium, STARmap, DBiT-seq, ATAC-seq, CODEX, IMC) to better understand how cells, tissues, organs and organisms connect structure and function. In the spirit of mapping biological complexity, 3D Spatial Omics combines high-resolution molecular profiling (e.g. transcriptomics, proteomics) with 3D spatial information within cells, tissues, organs, or even whole organisms, preserving the architectural and functional relationships between cells in their native 3D microenvironment.
Nowadays, cancer or brain-related diseases remain significant global health challenges, driven by complex interactions and heterogeneity within disease-local microenvironment, as well as effective remote organs connected by nervous system, vascular system or lymphatic system. Systematic studies aim to holistically understand pathogenesis, progression, and therapeutic responses by integrating multiscale-multimodality data and clinical data, moving beyond reductionist methodology to uncover network-level mechanisms driving diseases.
3D spatial omics techniques often generate datasets ranging from hundreds of gigabytes to petabytes in size for a single experiment. Traditional computational tools including open-source ones and commercialized ones, cannot handle such big 3D images in an efficient, accurate and scalable way. New algorithms, tools, software would be developed to address the multi-tasks challenges involving preprocessing, postprocessing and higher-order analysis of 3D spatial omics data from raw images. Furthermore, integrating with clinical data and other types of functional data, multiscale-multimodal dataset would be used to pretrain foundation models for biology across text, image and video.