ProtIR: Iterative Refinement between Retrievers and Predictors for Protein Function Annotation
This addresses the problem of accurate protein function annotation in biology, offering a novel hybrid approach that is incremental but effective for domain-specific applications.
The paper tackles protein function annotation by introducing ProtIR, a variational pseudo-likelihood framework that iteratively refines knowledge between predictors and retrievers to incorporate inter-protein similarity modeling, achieving around 10% improvement over vanilla predictor-based methods and performance competitive with protein language models without large-scale pre-training.
Protein function annotation is an important yet challenging task in biology. Recent deep learning advancements show significant potential for accurate function prediction by learning from protein sequences and structures. Nevertheless, these predictor-based methods often overlook the modeling of protein similarity, an idea commonly employed in traditional approaches using sequence or structure retrieval tools. To fill this gap, we first study the effect of inter-protein similarity modeling by benchmarking retriever-based methods against predictors on protein function annotation tasks. Our results show that retrievers can match or outperform predictors without large-scale pre-training. Building on these insights, we introduce a novel variational pseudo-likelihood framework, ProtIR, designed to improve function predictors by incorporating inter-protein similarity modeling. This framework iteratively refines knowledge between a function predictor and retriever, thereby combining the strengths of both predictors and retrievers. ProtIR showcases around 10% improvement over vanilla predictor-based methods. Besides, it achieves performance on par with protein language model-based methods, yet without the need for massive pre-training, highlighting the efficacy of our framework. Code will be released upon acceptance.