Pruning-Assisted Modeling of Network Graph Connectivity from Spatial Transcriptomic Data

Functional interactions within and between different types of somatic cells are crucial for executing complex organ-level biological processes in multicellular organisms. Spatial transcriptomic technologies have allowed for high throughput characterization of cell communities and associated cellular processes in the tissue contexts. However, analytical resources for characterization and quantitative inference of spatial interactions among somatic cells that can potentially impact complex biological functions in tissue microenvironment are still limited. Here, we describe a framework to use network graph-based spatial statistical models on spatially annotated molecular data to gain insights into cellular relationship and connectivity in the local tumor microenvironment and evaluate the effects of network graph connectivity on the model inference.