Leonardo Pacheco Londono

In reanalyzed PXD016245 monocyte-differentiation proteomes, eight predefined protein-module scores retain a joint condition association (incremental R^2 0.313663, permutation p = 0.0002) beyond source block and day.

The paper analyzes immune gene co-expression networks across two cohorts and finds that neither modularity nor spectral dominance alone distinguishes healthy from disease states, but their joint distribution reveals a geometric constraint.