A Transferability-Based Method for Evaluating the Protein Representation Learning

Self-supervised pre-trained language models have recently risen as a powerful approach in learning protein representations, showing exceptional effectiveness in various biological tasks, such as drug discovery. Amidst the evolving trend in protein language model development, there is an observable shift towards employing large-scale multimodal and multitask models. However, the predominant reliance on empirical assessments using specific benchmark datasets for evaluating these models raises concerns about the comprehensiveness and efficiency of current evaluation methods. Addressing this gap, our study introduces a novel quantitative approach for estimating the performance of transferring multi-task pre-trained protein representations to downstream tasks. This transferability-based method is designed to quantify the similarities in latent space distributions between pre-trained features and those fine-tuned for downstream tasks. It encompasses a broad spectrum, covering multiple domains and a variety of heterogeneous tasks. To validate this method, we constructed a diverse set of protein-specific pre-training tasks. The resulting protein representations were then evaluated across several downstream biological tasks. Our experimental results demonstrate a robust correlation between the transferability scores obtained using our method and the actual transfer performance observed. This significant correlation highlights the potential of our method as a more comprehensive and efficient tool for evaluating protein representation learning.

Authors

• Fan Hu
• Weihong Zhang Washington University; Nevin L. Zhang
• Huazhen Huang
• Wang Li
• Yang Li Northwestern Polytechnical University
• Peng Yin

Keywords

• Task analysis
• Proteins
• Protein engineering
• Biological system modeling
• Computational modeling
• Predictive models
• Biological information theory
• Representation Learning
• Protein Representation Learning
• Quantitative Data
• Latent Space
• Language Model
• Transfer Performance
• Empirical Assessment
• Multimodal Model
• Pre-trained Language Models
• Pre-training Tasks
• Gene Ontology
• Efficient Way
• Feature Information
• Feature Representation
• Gene Ontology Annotation
• Multiple Tasks
• Computational Biology
• Random Group
• Cell Biol
• Label Information
• Masked Language Model
• Wasserstein Distance
• Field Of Natural Language Processing
• Diverse Tasks
• Combined Objective
• Transfer Distance
• Fine-tuning Process
• Optimal Transport
• Text Classification
• Pre-training Phase
• Transferability
• Machine Learning
• Humans
• Databases, Protein
• Algorithms