Graph-based autoencoder integrates spatial transcriptomics with chromatin images and identifies joint biomarkers for Alzheimer’s disease

• DOI: 10.1038/s41467-022-35233-1
• Publication Date: 2022-12-03T15:48:34Z
• Authors: Xinyi Zhang; Xiao Wang; G. V. Shivashankar; Caroline Uhler
• Keywords: Autoencoder

Tissue development and disease lead to changes in cellular organization, nuclear morphology, gene expression, which can be jointly measured by spatial transcriptomic technologies. However, methods for analyzing the different data modalities 3D are still lacking. We present a computational framework integrate Spatial Transcriptomic using over-parameterized graph-based Autoencoders with Chromatin Imaging (STACI) identify molecular functional alterations tissues. STACI incorporates multiple single representation downstream tasks, enables prediction of from images unseen tissue sections, provides built-in batch correction expression morphology through over-parameterization. apply analyze spatio-temporal progression Alzheimer's associated morphometric coupled features. Collectively, we demonstrate importance characterizing integrating its potential discovery biomarkers.