Researchers at The University of Texas MD Anderson Cancer Center developed an AI tool called SPACER. It explains how immune cells navigate tissues and why they gather or avoid certain areas. SPACER uses spatial transcriptomics to analyze gene activity in intact tissue. It treats groups of neighboring cells as units rather than isolated entities. The tool was four to 21 times more effective than existing methods at identifying biologically meaningful signals. It found that immune cells respond primarily to local signals rather than body-wide ones. Tumors with active antigen presentation genes attract T cells more effectively. Mucins, sugar-coated proteins, may form barriers that keep T cells away from tumors. Higher mucin levels in melanoma and gastric cancer correlate with poorer immunotherapy responses. SPACER also identified signals linked to T cell recruitment in heart inflammation models. The tool could aid cancer therapy diagnostics and improve understanding of autoimmune diseases and tissue repair.

• 4- to 21-times more effective than existing approaches
• 37 datasets analyzed