DeepFold-PLM: Accelerating Protein Structure Prediction via Efficient Homology Search Using Protein Language Models
1 Department of Physics, Sungkyunkwan University, Suwon, Korea.
2 Basic Science Research Institute, Changwon National University, Changwon 51140, Korea
3 Center for Advanced Computation, Korea Institute for Advanced Study, Seoul, Korea
† These authors contributed equally to this work.
* Corresponding authors: jejoong@skku.edu (J. Yoo), newton@kias.re.kr (K. Joo)
Abstract
Protein structure prediction has been revolutionized and generalized with the advent of cutting-edge AI methods such as AlphaFold, but reliance on computationally intensive multiple sequence alignments (MSA) remains a major limitation. We introduce DeepFold-PLM, a novel framework that integrates advanced protein language models (PLMs) with vector embedding databases to enhance ultra-fast MSA construction, remote homology detection, and protein structure prediction. DeepFold-PLM utilizes high-dimensional embeddings and contrastive learning, significantly accelerate MSA generation, achieving 47 times faster than standard methods, while maintaining prediction accuracy comparable to AlphaFold. In addition, it enhances structure prediction by extending modeling capabilities to multimeric protein complexes, provides a scalable PyTorch-based implementation for efficient large-scale prediction, and offers a user-friendly web service for real-time analysis. Our method also effectively increases sequence diversity, enriching coevolutionary information critical for accurate structure prediction. DeepFold-PLM thus represents a versatile and practical resource that enables high-throughput applications in computational structural biology.
* This paper is not yet published. A link will be provided once it becomes available.