Conditionally Site-Independent Neural Evolution of Antibody Sequences

2026-02-21

AI Summary

This work addresses a critical limitation in current deep learning approaches to antibody engineering, which overlook the evolutionary dynamics of affinity maturation, while traditional phylogenetic models struggle to capture complex epistatic interactions among sites. To bridge this gap, the authors propose CoSiNE—a continuous-time Markov chain parameterized by deep neural networks—that uniquely integrates deep learning with continuous-time evolutionary modeling. CoSiNE explicitly disentangles somatic hypermutation from selection pressure and introduces a Guided Gillespie sampling strategy to optimize antigen-binding affinity. Theoretically, the model is shown to provide a first-order approximation of the sequence point mutation process. In zero-shot variant effect prediction tasks, CoSiNE outperforms state-of-the-art language models and efficiently generates high-affinity antibody sequences.

Technology Category

• Machine Learning: Evolutionary Learning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Models for Web evolution
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

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.

Problem

• antibody engineering
• affinity maturation
• epistatic interactions
• evolutionary dynamics
• sequence modeling

Innovation

• CoSiNE
• epistatic interactions
• affinity maturation
• continuous-time Markov chain
• Guided Gillespie