The landscape of plasma proteomic links to human organ imaging

Abstract

Plasma protein levels provide important insights into human disease, yet a comprehensive assessment of plasma proteomics across organs is lacking. Using large-scale multimodal data from the UK Biobank, we integrate plasma proteomics with organ imaging to map their phenotypic and genetic links, analyzing 2923 proteins and 1051 imaging traits across multiple organs. We uncover 5067 phenotypic protein-imaging associations, identifying both organ-specific and organ-shared proteomic relations, along with enriched protein-protein interaction networks and biological pathways. Sensitivity analyses suggest that these associations are not substantially influenced by the median 10.18-year interval between plasma sampling and imaging visits. We also map key protein predictors of organ structures and show the stratification capability of plasma protein-based prediction models. Furthermore, we identify 8116 putative causal protein-imaging links. Imaging-associated protein components show enrichment across diverse complex diseases. Our study shows that integrating plasma proteomics with multi-organ imaging provides a comprehensive pan-organ imaging-proteomics map and reveals molecular pathways linking circulating proteins to human organ biology. Fan and colleagues present a study where they map links between plasma proteins and imaging traits across multiple human organs, revealing organ-specific and cross-organ protein signatures, predictive biomarkers, and genetic evidence supporting causal protein-imaging relationships.