{"title":"基于扩散的新合成启动子设计生成网络","authors":"Jianfeng Lin, Xin Wang, Tuoyu Liu, Yue Teng* and Wei Cui*, ","doi":"10.1021/acssynbio.4c00041","DOIUrl":null,"url":null,"abstract":"<p >Computer-aided promoter design is a major development trend in synthetic promoter engineering. Various deep learning models have been used to evaluate or screen synthetic promoters, but there have been few works on de novo promoter design. To explore the potential ability of generative models in promoter design, we established a diffusion-based generative model for promoter design in <i>Escherichia coli</i>. The model was completely driven by sequence data and could study the essential characteristics of natural promoters, thus generating synthetic promoters similar to natural promoters in structure and component. We also improved the calculation method of FID indicator, using a convolution layer to extract the feature matrix of the promoter sequence instead. As a result, we got an FID equal to 1.37, which meant synthetic promoters have a distribution similar to that of natural ones. Our work provides a fresh approach to de novo promoter design, indicating that a completely data-driven generative model is feasible for promoter design.</p>","PeriodicalId":26,"journal":{"name":"ACS Synthetic Biology","volume":null,"pages":null},"PeriodicalIF":3.7000,"publicationDate":"2024-04-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Diffusion-Based Generative Network for de Novo Synthetic Promoter Design\",\"authors\":\"Jianfeng Lin, Xin Wang, Tuoyu Liu, Yue Teng* and Wei Cui*, \",\"doi\":\"10.1021/acssynbio.4c00041\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p >Computer-aided promoter design is a major development trend in synthetic promoter engineering. Various deep learning models have been used to evaluate or screen synthetic promoters, but there have been few works on de novo promoter design. To explore the potential ability of generative models in promoter design, we established a diffusion-based generative model for promoter design in <i>Escherichia coli</i>. The model was completely driven by sequence data and could study the essential characteristics of natural promoters, thus generating synthetic promoters similar to natural promoters in structure and component. We also improved the calculation method of FID indicator, using a convolution layer to extract the feature matrix of the promoter sequence instead. As a result, we got an FID equal to 1.37, which meant synthetic promoters have a distribution similar to that of natural ones. Our work provides a fresh approach to de novo promoter design, indicating that a completely data-driven generative model is feasible for promoter design.</p>\",\"PeriodicalId\":26,\"journal\":{\"name\":\"ACS Synthetic Biology\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":3.7000,\"publicationDate\":\"2024-04-13\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"ACS Synthetic Biology\",\"FirstCategoryId\":\"99\",\"ListUrlMain\":\"https://pubs.acs.org/doi/10.1021/acssynbio.4c00041\",\"RegionNum\":2,\"RegionCategory\":\"生物学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"BIOCHEMICAL RESEARCH METHODS\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"ACS Synthetic Biology","FirstCategoryId":"99","ListUrlMain":"https://pubs.acs.org/doi/10.1021/acssynbio.4c00041","RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"BIOCHEMICAL RESEARCH METHODS","Score":null,"Total":0}
Diffusion-Based Generative Network for de Novo Synthetic Promoter Design
Computer-aided promoter design is a major development trend in synthetic promoter engineering. Various deep learning models have been used to evaluate or screen synthetic promoters, but there have been few works on de novo promoter design. To explore the potential ability of generative models in promoter design, we established a diffusion-based generative model for promoter design in Escherichia coli. The model was completely driven by sequence data and could study the essential characteristics of natural promoters, thus generating synthetic promoters similar to natural promoters in structure and component. We also improved the calculation method of FID indicator, using a convolution layer to extract the feature matrix of the promoter sequence instead. As a result, we got an FID equal to 1.37, which meant synthetic promoters have a distribution similar to that of natural ones. Our work provides a fresh approach to de novo promoter design, indicating that a completely data-driven generative model is feasible for promoter design.
期刊介绍:
The journal is particularly interested in studies on the design and synthesis of new genetic circuits and gene products; computational methods in the design of systems; and integrative applied approaches to understanding disease and metabolism.
Topics may include, but are not limited to:
Design and optimization of genetic systems
Genetic circuit design and their principles for their organization into programs
Computational methods to aid the design of genetic systems
Experimental methods to quantify genetic parts, circuits, and metabolic fluxes
Genetic parts libraries: their creation, analysis, and ontological representation
Protein engineering including computational design
Metabolic engineering and cellular manufacturing, including biomass conversion
Natural product access, engineering, and production
Creative and innovative applications of cellular programming
Medical applications, tissue engineering, and the programming of therapeutic cells
Minimal cell design and construction
Genomics and genome replacement strategies
Viral engineering
Automated and robotic assembly platforms for synthetic biology
DNA synthesis methodologies
Metagenomics and synthetic metagenomic analysis
Bioinformatics applied to gene discovery, chemoinformatics, and pathway construction
Gene optimization
Methods for genome-scale measurements of transcription and metabolomics
Systems biology and methods to integrate multiple data sources
in vitro and cell-free synthetic biology and molecular programming
Nucleic acid engineering.