Protein Language Models in Virology: A Review of Advances and Applications

Protein language models (PLMs) enable functional analysis of divergent viral sequences without homology alignment. This review covers PLM architectures from sequence encoders through structure-aware architectures to generative models and assesses their application to orphan protein structure resolution, virosphere-wide functional classification, host factor identification, and therapeutic antibody optimization. Finally, the limitations of the current model in terms of interpretability and insufficient data representation are discussed while exploring future trends toward multimodal integration and the "dry-wet" experimental loop to accelerate the adoption of artificial intelligence in precision virology.