Profiling adaptive immunity: A quantitative framework for immune repertoire dynamics and clinical diagnostics

While immune repertoire (IR) dynamics encode critical disease signatures, existing models fail to capture the non-equilibrium processes that govern clonal selection and somatic hypermutation (SHM). We develop a biophysical framework that mathematically reconstructs IR evolution through energy landscape optimization, where 1) clonal emergence probabilities map to metastable states, 2) repertoire transitions obey non-equilibrium dynamics, and 3) inter-IR distances quantify the distribution transformation costs via optimal transport theory. This model enables macroscopic immune state detection from 10k cells by resolving critical fluctuations in sparse sampling regimes. Experimental validation across murine/human cohorts demonstrates precise unsupervised stratification of immune stages and disease states without prior clinical annotations. By bridging stochastic SHM kinetics with deterministic repertoire shifts, our approach establishes quantitative metrics for tracking immunological trajectories and pathological progression. This paradigm shift from static pattern recognition to dynamic process decoding empowers sensitive immunomonitoring and personalized therapeutic design through ultra-low-input clinical sampling.