TSNR: Protein-Specific Signals with Broad Molecular Context for Spatial Protein Prediction

TSNR, a spatial protein prediction method on the Spatial-CITE-seq benchmark, achieves the highest mean macro Spearman correlation among reproduced methods under matched biological holdout evaluation.

• Why it matters: Spatial transcriptomics can infer protein abundance beyond measured antibody panels, but it is unclear which molecular context to keep and how well it transfers across tissue environments.
• What they did: The authors compare predictors sharing a protein-specific center representation while varying contextual molecular scope, then analyze value-mismatch and estimating-score variation; they propose Target-Specific Niche Regression using a protein-specific Huber regressor.
• The result: TSNR’s broader context plus bounded-score robust regression is supported by controlled analyses, and in two external cohorts protein-level gains are positively rank-correlated with the value-mismatch balance across 11 held-out biological folds.

AI summary of the abstract · may contain errors · not medical advice · read the paper.