Joseph Boen

This study uses spatial multi-omics and deep learning to identify immunotherapy response and resistance signatures in hepatocellular carcinoma.

The paper introduces CORAL, a deep generative model for spatial multiomics data that deconvolves low-resolution data into single-cell profiles and identifies spatial domains and cell-cell interactions.

The paper develops spatial descriptors using mathematical tools to analyze cell interactions in colorectal cancer progression and demonstrates their utility in identifying disease-related pathways.

This work infers copy number variations from scRNA-seq data using three methods, compares them statistically, and applies the results to analyze tumor cell subpopulations and heterogeneity.