Comprehensive Analysis of SARS-CoV-2 Spike Evolution: Epitope Classification and Immune Escape Prediction

The evolution of SARS-CoV-2, the virus responsible for the COVID-19 pandemic, has produced unprecedented numbers of structures of the Spike protein. In this study, we present a comprehensive analysis of 1,560 published structures, covering most major variants that emerged throughout the pandemic, diverse heteromerization and interacting complexes. Using interaction-energy-informed geometric clustering, we identify 14 structurally distinct epitopes based on their conformational specificity, shared interface with ACE2, and glycosylation patterns. Our per-residue interaction evaluations accurately predict antibody recognition sites and correlate strongly with deep mutational scanning (DMS) data, enabling immune escape predictions for future variants. To complement this structural analysis, we integrate longitudinal genomic data from nearly 3 million viral sequences, linking mutational patterns to changes in Spike’s conformational dynamics. Our findings reveal two distinct evolutionary trade-offs driving immune escape. First, we confirm an enthalpic trade-off, where mutations in the receptor-binding motif (RBM) enhance immune escape at the cost of weakened ACE2 binding. Second, we introduce an entropic trade-off, showing that mutations outside the RBM modulate Spike’s conformational equilibrium, reducing open-state occupancy to evade immune detection—without directly altering the ACE2-binding interface. With these analyses, this work not only highlights the different functional effects of mutations across SARS-CoV-2 Spike variants but also reveals the complex interplay of evolutionary forces shaping the evolution of the SARS-CoV-2 Spike protein over the course of the pandemic.