LightRoseTTA: High-efficient and Accurate Protein Structure Prediction Using an Ultra-Lightweight Deep Graph Model

Abstract

AbstractAccurately predicting protein structure, from amino acid sequences to three-dimensional structures, is of great significance in biological research. To tackle this issue, a representative deep big model, RoseTTAFold, has been proposed with promising success. Here, we reportan ultra-lightweight deep graph network, namedLightRoseTTA, to achieve accurate and high-efficient prediction for proteins. Notably, three highlights are possessed by our LightRoseTTA:(i) high-accuratestructure prediction for proteins, beingcompetitive with RoseTTAFoldon multiple popular datasets including CASP14 and CAMEO;(ii) high-efficienttraining and inference with an ultra-lightweight model, costingonly one week on one single general NVIDIA 3090 GPU for model-training(vs 30 days on 8 high-speed NVIDIA V100 GPUs for RoseTTAFold) and containingonly 1.4M parameters(vs 130M in RoseTTAFold);(iii) low dependencyon multi-sequence alignments (MSA, widely-used homologous information), achievingthe best performance on three MSA-insufficient datasets: Orphan, De novo, and Orphan25. Besides, our LightRoseTTA istransferablefrom general proteins to antibody data, as verified in our experiments. We visualize some case studies to demonstrate the high-quality prediction, and provide some insights on how the structure predictions facilitate the understanding of biological functions. We further make a discussion on the time and resource costs of LightRoseTTA and RoseTTAFold, and demonstrate the feasibility of lightweight models for protein structure prediction, which may be crucial in the resource-limited research for universities and academy institutions.We release our code and model to speed biological research.