Multi-Modal CLIP-Informed Protein Editing

Background: Proteins govern most biological functions essential for life, and achieving controllable protein editing has made great advances in probing natural systems, creating therapeutic conjugates, and generating novel protein constructs. Recently, machine learning-assisted protein editing (MLPE...

Full description

Saved in:
Bibliographic Details
Main Authors: Mingze Yin, Hanjing Zhou, Yiheng Zhu, Miao Lin, Yixuan Wu, Jialu Wu, Hongxia Xu, Chang-Yu Hsieh, Tingjun Hou, Jintai Chen, Jian Wu
Format: Article
Language:English
Published: American Association for the Advancement of Science (AAAS) 2024-01-01
Series:Health Data Science
Online Access:https://spj.science.org/doi/10.34133/hds.0211
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846115023726837760
author Mingze Yin
Hanjing Zhou
Yiheng Zhu
Miao Lin
Yixuan Wu
Jialu Wu
Hongxia Xu
Chang-Yu Hsieh
Tingjun Hou
Jintai Chen
Jian Wu
author_facet Mingze Yin
Hanjing Zhou
Yiheng Zhu
Miao Lin
Yixuan Wu
Jialu Wu
Hongxia Xu
Chang-Yu Hsieh
Tingjun Hou
Jintai Chen
Jian Wu
author_sort Mingze Yin
collection DOAJ
description Background: Proteins govern most biological functions essential for life, and achieving controllable protein editing has made great advances in probing natural systems, creating therapeutic conjugates, and generating novel protein constructs. Recently, machine learning-assisted protein editing (MLPE) has shown promise in accelerating optimization cycles and reducing experimental workloads. However, current methods struggle with the vast combinatorial space of potential protein edits and cannot explicitly conduct protein editing using biotext instructions, limiting their interactivity with human feedback. Methods: To fill these gaps, we propose a novel method called ProtET for efficient CLIP-informed protein editing through multi-modality learning. Our approach comprises 2 stages: In the pretraining stage, contrastive learning aligns protein–biotext representations encoded by 2 large language models (LLMs). Subsequently, during the protein editing stage, the fused features from editing instruction texts and original protein sequences serve as the final editing condition for generating target protein sequences. Results: Comprehensive experiments demonstrated the superiority of ProtET in editing proteins to enhance human-expected functionality across multiple attribute domains, including enzyme catalytic activity, protein stability, and antibody-specific binding ability. ProtET improves the state-of-the-art results by a large margin, leading to substantial stability improvements of 16.67% and 16.90%. Conclusions: This capability positions ProtET to advance real-world artificial protein editing, potentially addressing unmet academic, industrial, and clinical needs.
format Article
id doaj-art-5d076180547f4fe1a9d009c80353247f
institution Kabale University
issn 2765-8783
language English
publishDate 2024-01-01
publisher American Association for the Advancement of Science (AAAS)
record_format Article
series Health Data Science
spelling doaj-art-5d076180547f4fe1a9d009c80353247f2024-12-20T01:44:45ZengAmerican Association for the Advancement of Science (AAAS)Health Data Science2765-87832024-01-01410.34133/hds.0211Multi-Modal CLIP-Informed Protein EditingMingze Yin0Hanjing Zhou1Yiheng Zhu2Miao Lin3Yixuan Wu4Jialu Wu5Hongxia Xu6Chang-Yu Hsieh7Tingjun Hou8Jintai Chen9Jian Wu10School of Medicine, Zhejiang University, Hangzhou, China.College of Computer Science and Technology, Zhejiang University, Hangzhou, China.College of Computer Science and Technology, Zhejiang University, Hangzhou, China.Medical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical University, Guangzhou, China.School of Medicine, Zhejiang University, Hangzhou, China.Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China.Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China.Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China.Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University, College of Pharmaceutical Sciences, Zhejiang University, Hangzhou, China.AI Thrust, Information Hub, HKUST (Guangzhou), Guangzhou, China.Second Affiliated Hospital School of Medicine, Hangzhou, China.Background: Proteins govern most biological functions essential for life, and achieving controllable protein editing has made great advances in probing natural systems, creating therapeutic conjugates, and generating novel protein constructs. Recently, machine learning-assisted protein editing (MLPE) has shown promise in accelerating optimization cycles and reducing experimental workloads. However, current methods struggle with the vast combinatorial space of potential protein edits and cannot explicitly conduct protein editing using biotext instructions, limiting their interactivity with human feedback. Methods: To fill these gaps, we propose a novel method called ProtET for efficient CLIP-informed protein editing through multi-modality learning. Our approach comprises 2 stages: In the pretraining stage, contrastive learning aligns protein–biotext representations encoded by 2 large language models (LLMs). Subsequently, during the protein editing stage, the fused features from editing instruction texts and original protein sequences serve as the final editing condition for generating target protein sequences. Results: Comprehensive experiments demonstrated the superiority of ProtET in editing proteins to enhance human-expected functionality across multiple attribute domains, including enzyme catalytic activity, protein stability, and antibody-specific binding ability. ProtET improves the state-of-the-art results by a large margin, leading to substantial stability improvements of 16.67% and 16.90%. Conclusions: This capability positions ProtET to advance real-world artificial protein editing, potentially addressing unmet academic, industrial, and clinical needs.https://spj.science.org/doi/10.34133/hds.0211
spellingShingle Mingze Yin
Hanjing Zhou
Yiheng Zhu
Miao Lin
Yixuan Wu
Jialu Wu
Hongxia Xu
Chang-Yu Hsieh
Tingjun Hou
Jintai Chen
Jian Wu
Multi-Modal CLIP-Informed Protein Editing
Health Data Science
title Multi-Modal CLIP-Informed Protein Editing
title_full Multi-Modal CLIP-Informed Protein Editing
title_fullStr Multi-Modal CLIP-Informed Protein Editing
title_full_unstemmed Multi-Modal CLIP-Informed Protein Editing
title_short Multi-Modal CLIP-Informed Protein Editing
title_sort multi modal clip informed protein editing
url https://spj.science.org/doi/10.34133/hds.0211
work_keys_str_mv AT mingzeyin multimodalclipinformedproteinediting
AT hanjingzhou multimodalclipinformedproteinediting
AT yihengzhu multimodalclipinformedproteinediting
AT miaolin multimodalclipinformedproteinediting
AT yixuanwu multimodalclipinformedproteinediting
AT jialuwu multimodalclipinformedproteinediting
AT hongxiaxu multimodalclipinformedproteinediting
AT changyuhsieh multimodalclipinformedproteinediting
AT tingjunhou multimodalclipinformedproteinediting
AT jintaichen multimodalclipinformedproteinediting
AT jianwu multimodalclipinformedproteinediting