Multiscale Multimodal Integration of Spatial Omics

In our work, we aim to investigate multiscale bone adaptation and regeneration using correlative multimodal imaging. Using in vivo 3D micro-computed tomography (micro-CT) images, ex vivo 2D histological sections and 2D spatial transcriptomics data from the same bone sample, we aim to connect structural and functional information from organ and tissue levels with biological data at the cellular and gene scales. Additionally, using micro-finite element analysis (micro-FE) on 3D micro-CT images, the local in vivo mechanical environment can be recreated and correlated with the cellular and gene data from 2D modalities, enabling a unique association of organ-to-cell events.

We are developing computational methods to preprocess 2D and 3D images with high-throughput, based on recent advances in deep-learning methods for image analysis, accurately register 2D and 3D complementary modalities of the same speciment and perform multimodal data integration. Our goal is to provide an end-to-end software that can support the multimodal analysis of multiscale systems focusing on bone applications and beyond.