LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model

Abstract Accurately predicting protein structure, from sequences to 3D structures, is of great significance in biological research. To tackle this issue, a representative deep big model, RoseTTAFold, is proposed with promising success. Here, “a light‐weight deep graph network, named LightRoseTTA,” i...

Full description

Saved in:
Bibliographic Details
Main Authors: Xudong Wang, Tong Zhang, Guangbu Liu, Zhen Cui, Zhiyong Zeng, Cheng Long, Wenming Zheng, Jian Yang
Format: Article
Language:English
Published: Wiley 2025-05-01
Series:Advanced Science
Subjects:
Online Access:https://doi.org/10.1002/advs.202309051
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850151397658460160
author Xudong Wang
Tong Zhang
Guangbu Liu
Zhen Cui
Zhiyong Zeng
Cheng Long
Wenming Zheng
Jian Yang
author_facet Xudong Wang
Tong Zhang
Guangbu Liu
Zhen Cui
Zhiyong Zeng
Cheng Long
Wenming Zheng
Jian Yang
author_sort Xudong Wang
collection DOAJ
description Abstract Accurately predicting protein structure, from sequences to 3D structures, is of great significance in biological research. To tackle this issue, a representative deep big model, RoseTTAFold, is proposed with promising success. Here, “a light‐weight deep graph network, named LightRoseTTA,” is reported to achieve accurate and highly efficient prediction for proteins. Notably, three highlights are possessed by LightRoseTTA: i) high‐accurate structure prediction for proteins, being “competitive with RoseTTAFold” on multiple popular datasets including CASP14 and CAMEO; ii) high‐efficient training and inference with a light‐weight model, costing “only 1 week on one single NVIDIA 3090 GPU for model‐training” (vs 30 days on 8 NVIDIA V100 GPUs for RoseTTAFold) and containing “only 1.4M parameters” (vs 130M in RoseTTAFold); iii) low dependency on multi‐sequence alignment (MSA), achieving the best performance on three MSA‐insufficient datasets: Orphan, De novo, and Orphan25. Besides, LightRoseTTA is “transferable” from general proteins to antibody data, as verified in the experiments. The time and resource costs of LightRoseTTA and RoseTTAFold are further discussed to demonstrate the feasibility of light‐weight models for protein structure prediction, which may be crucial in resource‐limited research for universities and academic institutions. The code and model are released to speed biological research (https://github.com/psp3dcg/LightRoseTTA).
format Article
id doaj-art-cca17df6818d4240a5fedea85b953922
institution OA Journals
issn 2198-3844
language English
publishDate 2025-05-01
publisher Wiley
record_format Article
series Advanced Science
spelling doaj-art-cca17df6818d4240a5fedea85b9539222025-08-20T02:26:18ZengWileyAdvanced Science2198-38442025-05-011219n/an/a10.1002/advs.202309051LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph ModelXudong Wang0Tong Zhang1Guangbu Liu2Zhen Cui3Zhiyong Zeng4Cheng Long5Wenming Zheng6Jian Yang7School of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 ChinaSchool of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 ChinaSchool of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 ChinaSchool of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 ChinaSchool of Automation Nanjing University of Science and Technology Nanjing 210094 ChinaSchool of Computer Engineering Nanyang Technological University No. 50, Nanyang Avenue Singapore 639798 SingaporeSchool of Biological Science & Medical Engineering Southeast University Nanjing 210096 ChinaSchool of Computer Science and Engineering Nanjing University of Science and Technology Nanjing 210094 ChinaAbstract Accurately predicting protein structure, from sequences to 3D structures, is of great significance in biological research. To tackle this issue, a representative deep big model, RoseTTAFold, is proposed with promising success. Here, “a light‐weight deep graph network, named LightRoseTTA,” is reported to achieve accurate and highly efficient prediction for proteins. Notably, three highlights are possessed by LightRoseTTA: i) high‐accurate structure prediction for proteins, being “competitive with RoseTTAFold” on multiple popular datasets including CASP14 and CAMEO; ii) high‐efficient training and inference with a light‐weight model, costing “only 1 week on one single NVIDIA 3090 GPU for model‐training” (vs 30 days on 8 NVIDIA V100 GPUs for RoseTTAFold) and containing “only 1.4M parameters” (vs 130M in RoseTTAFold); iii) low dependency on multi‐sequence alignment (MSA), achieving the best performance on three MSA‐insufficient datasets: Orphan, De novo, and Orphan25. Besides, LightRoseTTA is “transferable” from general proteins to antibody data, as verified in the experiments. The time and resource costs of LightRoseTTA and RoseTTAFold are further discussed to demonstrate the feasibility of light‐weight models for protein structure prediction, which may be crucial in resource‐limited research for universities and academic institutions. The code and model are released to speed biological research (https://github.com/psp3dcg/LightRoseTTA).https://doi.org/10.1002/advs.202309051graph neural networklight‐weight deep learning modelprotein structure prediction
spellingShingle Xudong Wang
Tong Zhang
Guangbu Liu
Zhen Cui
Zhiyong Zeng
Cheng Long
Wenming Zheng
Jian Yang
LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
Advanced Science
graph neural network
light‐weight deep learning model
protein structure prediction
title LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
title_full LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
title_fullStr LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
title_full_unstemmed LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
title_short LightRoseTTA: High‐Efficient and Accurate Protein Structure Prediction Using a Light‐Weight Deep Graph Model
title_sort lightrosetta high efficient and accurate protein structure prediction using a light weight deep graph model
topic graph neural network
light‐weight deep learning model
protein structure prediction
url https://doi.org/10.1002/advs.202309051
work_keys_str_mv AT xudongwang lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT tongzhang lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT guangbuliu lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT zhencui lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT zhiyongzeng lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT chenglong lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT wenmingzheng lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel
AT jianyang lightrosettahighefficientandaccurateproteinstructurepredictionusingalightweightdeepgraphmodel