Protocol for identifying cellular and molecular signatures from spatial profiling data using micro-tissue analysis

Spatial tissue profiling measures hundreds to thousands of biomolecules across tissue regions, but most analyses examine each feature one at a time. Here, we present a protocol for micro-tissue analysis using partial least-squares modeling. We describe steps for stratifying regions by microenvironment, building multivariate models to identify which features collectively drive a biological outcome, and interpreting the results. We detail four worked examples using two published GeoMx digital spatial profiler (DSP) datasets from human tumor and mucosal tissues. For complete details on the use and execution of this protocol, please refer to Lu et al.1 and Ng et al.2.