Conditionally Site-Independent Neural Evolution of Antibody Sequences

Common deep learning approaches for antibody engineering focus on modeling the marginal distribution of sequences. By treating sequences as independent samples, however, these methods overlook affinity maturation as a rich and largely untapped source of information about the evolutionary process by which antibodies explore the underlying fitness landscape. In contrast, classical phylogenetic models explicitly represent evolutionary dynamics but lack the expressivity to capture complex epistatic interactions. We bridge this gap with CoSiNE, a continuous-time Markov chain parameterized by a deep neural network. Mathematically, we prove that CoSiNE provides a first-order approximation to the intractable sequential point mutation process, capturing epistatic effects with an error bound that is quadratic in branch length. Empirically, CoSiNE outperforms state-of-the-art language models in zero-shot variant effect prediction by explicitly disentangling selection from context-dependent somatic hypermutation. Finally, we introduce Guided Gillespie, a classifier-guided sampling scheme that steers CoSiNE at inference time, enabling efficient optimization of antibody binding affinity toward specific antigens.

When your immune system fights an infection, antibodies inside your body undergo a rapid evolution called affinity maturation: they mutate repeatedly, and the variants that bind the invader best are selected for further refinement. Understanding this process could help us design better therapeutic antibodies, but capturing it has been difficult. Modern AI methods treat antibody sequences as independent snapshots and ignore the evolutionary process that produced them. Classical evolutionary models do represent this process, but assume that each position in the protein mutates independently of every other position. This misses the reality that amino acids in an antibody work together to recognize their target. We developed CoSiNE, a method that uses a neural network to predict how an antibody is likely to evolve next, while accounting for interactions between positions. We prove mathematically that this approximation is accurate, and we provide a sampling procedure that steers the model toward antibodies that bind specific targets such as viruses. CoSiNE achieves state-of-the-art performance at predicting which antibody mutations improve function, providing a new tool for antibody design grounded in the biology of how antibodies actually evolve.