DeepProtein: Deep Learning Library and Benchmark for Protein Sequence Learning

TL;DR

DeepProtein is a new deep learning library and benchmark suite designed for protein sequence analysis, enabling researchers to evaluate various models across multiple protein-related tasks with state-of-the-art results.

Contribution

It introduces a comprehensive library and benchmark for protein tasks, along with DeepProt-T5 models that achieve top performance, advancing protein sequence learning research.

Findings

DeepProt-T5 models achieve state-of-the-art results on four tasks.

Benchmark evaluates multiple architectures across diverse protein tasks.

DeepProtein is publicly available with extensive documentation.

Abstract

Deep learning has deeply influenced protein science, enabling breakthroughs in predicting protein properties, higher-order structures, and molecular interactions. This paper introduces DeepProtein, a comprehensive and user-friendly deep learning library tailored for protein-related tasks. It enables researchers to seamlessly address protein data with cutting-edge deep learning models. To assess model performance, we establish a benchmark evaluating different deep learning architectures across multiple protein-related tasks, including protein function prediction, subcellular localization prediction, protein-protein interaction prediction, and protein structure prediction. Furthermore, we introduce DeepProt-T5, a series of fine-tuned Prot-T5-based models that achieve state-of-the-art performance on four benchmark tasks, while demonstrating competitive results on six of others. Comprehensive documentation and tutorials are available which could ensure accessibility and support reproducibility. Built upon the widely used drug discovery library DeepPurpose, DeepProtein is publicly available at https://github.com/jiaqingxie/DeepProtein.