AbstractTranscriptional enhancers act as docking stations for combinations of transcription factors (TFs) and thereby regulate spatiotemporal activation of their target genes. It has been a long-standing goal in the field to decode the regulatory logic of an enhancer and to understand the details of how spatiotemporal gene expression is encoded in an enhancer sequence.

摘要转录增强子作为转录因子（TFs）组合的停靠站，从而调节其靶基因的时空激活。解码增强子的调控逻辑并了解增强子序列中如何编码时空基因表达的细节一直是该领域的长期目标。

Here, we show that deep learning models can be used to efficiently design synthetic, cell type specific enhancers, starting from random sequences, and that this optimization process allows for a detailed tracing of enhancer features at single-nucleotide resolution. We evaluate the function of fully synthetic enhancers to specifically target Kenyon cells or glial cells in the fruit fly brain using transgenic animals.

在这里，我们表明，深度学习模型可用于从随机序列开始有效设计合成的细胞类型特异性增强子，并且该优化过程允许以单核苷酸分辨率详细追踪增强子特征。我们使用转基因动物评估了全合成增强子特异性靶向果蝇大脑中的Kenyon细胞或神经胶质细胞的功能。

We further exploit enhancer design to create “dual-code” enhancers that target two cell types, and minimal enhancers smaller than 50 base pairs that are fully functional. By examining the state space searches towards local optima, we characterise enhancer codes through the strength, combination, and arrangement of TF activator and TF repressor motifs.

我们进一步利用增强子设计来创建针对两种细胞类型的“双代码”增强子，以及小于50个碱基对且功能齐全的最小增强子。通过检查状态空间对局部最优的搜索，我们通过TF激活子和TF阻遏子基序的强度，组合和排列来表征增强子代码。

Finally, we apply the same strategies to successfully design human enhancers, which adhere to similar enhancer rules as Drosophila enhancers. Enhancer design guided by deep learning leads to better understanding of how enhancers work and shows that their code can be exploited to manipulate cell states..

最后，我们应用相同的策略成功设计了人类增强子，这些增强子遵循与果蝇增强子相似的增强子规则。由深度学习指导的增强子设计可以更好地理解增强子的工作原理，并表明可以利用它们的代码来操纵细胞状态。。

You have full access to this article via your institution.

您可以通过您的机构完全访问本文。

Download PDF

下载PDF

Download PDF

下载PDF

Author informationAuthor notesJoy N. IsmailPresent address: UK Dementia Research Institute at Imperial College London, London, UKAuthors and AffiliationsLaboratory of Computational Biology, VIB Center for AI & Computational Biology (VIB.AI), Leuven, BelgiumIbrahim I. Taskiran, Katina I. Spanier, Hannah Dickmänken, Niklas Kempynck, Alexandra Pančíková, Eren Can Ekşi, Gert Hulselmans, Joy N.

作者信息作者notesJoy N.Ismail目前的地址：英国伦敦帝国理工学院英国痴呆症研究所，英国作者和附属机构计算生物学实验室，人工智能与计算生物学VIB中心（VIB.AI），鲁汶，BelgiumIbrahim I.Taskiran，Katina I.Spanier，Hannah Dickmänken，Niklas Kempynck，Alexandra Pančíková，Eren Can EkşI，Gert Hulselmans，乔伊N。

Ismail, Koen Theunis, Roel Vandepoel, Valerie Christiaens, David Mauduit & Stein AertsVIB-KULeuven Center for Brain & Disease Research, Leuven, BelgiumIbrahim I. Taskiran, Katina I. Spanier, Hannah Dickmänken, Niklas Kempynck, Alexandra Pančíková, Eren Can Ekşi, Gert Hulselmans, Koen Theunis, Roel Vandepoel, Valerie Christiaens, David Mauduit & Stein AertsDepartment of Human Genetics, KU Leuven, Leuven, BelgiumIbrahim I.

伊斯梅尔（Ismail）、科恩·特尤尼斯（Koen Theunis）、罗尔·范德波尔（Roel Vandepoel）、瓦莱丽·克里斯蒂安（Valerie Christiaens）、戴维·莫杜伊特（David Mauduit）和斯坦·阿尔茨维布·库鲁汶（Stein AertsVIB KULeuven）大脑与疾病研究中心（Center for Brain&Disease Research）、鲁汶（Leuven）、贝柳米·易卜拉欣（BelgiumIbrahim I.Taskiran）、卡蒂娜·斯潘尼尔（Katina I.Spanier）、汉娜·迪肯（Hannah Dickmänken）、尼古拉斯·凯，David Mauduit＆Stein Aerts鲁汶大学人类遗传学系，鲁汶，BelgiumIbrahim I。

Taskiran, Katina I. Spanier, Hannah Dickmänken, Niklas Kempynck, Alexandra Pančíková, Eren Can Ekşi, Gert Hulselmans, Joy N. Ismail, Koen Theunis, Roel Vandepoel, Valerie Christiaens, David Mauduit & Stein AertsVIB-KULeuven Center for Cancer Biology, Leuven, BelgiumAlexandra PančíkováAuthorsIbrahim I.

Taskiran、Katina I.Spanier、Hannah Dickmänken、Niklas Kempynck、Alexandra Pančíková、Eren Can EkšI、Gert Hulselmans、Joy N.Ismail、Koen Theunis、Roel Vandepel、Valerie Christiaens、David Mauduit Stein AertsVIB-KULeuven癌症生物学中心，Leuven，比利时Leuven。

TaskiranView author publicationsYou can also search for this author in.

TaskiranView作者出版物您也可以在中搜索此作者。

PubMed Google ScholarKatina I. SpanierView author publicationsYou can also search for this author in

PubMed Google ScholarKatina I.SpanierView作者出版物您也可以在

PubMed Google ScholarHannah DickmänkenView author publicationsYou can also search for this author in

PubMed Google ScholarHannah DickmänkenView作者出版物您也可以在

PubMed Google ScholarNiklas KempynckView author publicationsYou can also search for this author in

PubMed Google ScholarNiklas KempynckView作者出版物您也可以在

PubMed Google ScholarAlexandra PančíkováView author publicationsYou can also search for this author in

PubMed Google Scholar Alexandra Pančíková查看作者出版物您也可以在中搜索此作者

PubMed Google ScholarEren Can EkşiView author publicationsYou can also search for this author in

PubMed Google ScholarEren可以查看作者出版物您也可以在

PubMed Google ScholarGert HulselmansView author publicationsYou can also search for this author in

PubMed Google ScholarGert HulselmansView作者出版物您也可以在

PubMed Google ScholarJoy N. IsmailView author publicationsYou can also search for this author in

PubMed Google ScholarJoy N.IsmailView作者出版物您也可以在

PubMed Google ScholarKoen TheunisView author publicationsYou can also search for this author in

PubMed Google ScholarKoen TheunisView作者出版物您也可以在

PubMed Google ScholarRoel VandepoelView author publicationsYou can also search for this author in

PubMed Google ScholarRoel VandepoelView作者出版物您也可以在

PubMed Google ScholarValerie ChristiaensView author publicationsYou can also search for this author in

PubMed谷歌学者Valerie ChristiaensView作者出版物您也可以在

PubMed Google ScholarDavid MauduitView author publicationsYou can also search for this author in

PubMed Google ScholarDavid MauduitView作者出版物您也可以在

PubMed Google ScholarStein AertsView author publicationsYou can also search for this author in

PubMed Google ScholarStein AertsView作者出版物您也可以在

PubMed Google ScholarCorresponding authorCorrespondence to

PubMed谷歌学者通讯社

Stein Aerts.Supplementary informationSupplementary InformationThis file contains Supplementary Notes and Supplementary Figures. Supplementary Notes: Notes about validation of motif level explainability of deep learning models, enhancer design by Generative Adversarial Networks, and assessment of the activator/repressor balance in human enhancers.

斯坦·阿尔茨。补充信息补充信息该文件包含补充说明和补充数字。补充说明：关于验证深度学习模型的基序水平可解释性，通过生成性对抗网络进行增强子设计以及评估人类增强子中激活剂/阻遏物平衡的说明。

Supplementary Figures 1-16: Nucleotide contribution scores of all sequences not already displayed in main or supplementary accompanied eventually of in silico saturation mutagenesis prediction scores. Plots of prediction scores for the selection of mutations for the dual-code enhancers.Reporting SummarySupplementary TablesSequences of all cloned Dropsophila and human enhancers.Supplementary CodeAll scripts used to design sequences, analyse data, and reproduce figures.Peer Review FileRights and permissionsReprints and PermissionsAbout this articleCite this articleTaskiran, I.I., Spanier, K.I., Dickmänken, H.

补充图1-16：所有序列的核苷酸贡献分数尚未在主要或补充中显示，最终伴随着计算机饱和诱变预测分数。用于选择双代码增强子突变的预测分数图。报告所有克隆的Dropsophila和人类增强子的摘要补充表序列。补充代码用于设计序列，分析数据和再现数字的所有脚本。同行评审文件权限和许可本文引用本文TaskIran，I.I.，Spanier，K.I.，Dickmänken，H。

et al. Cell type directed design of synthetic enhancers..

等人。合成增强子的细胞类型导向设计。。

Nature (2023). https://doi.org/10.1038/s41586-023-06936-2Download citationReceived: 06 July 2022Accepted: 05 December 2023Published: 12 December 2023DOI: https://doi.org/10.1038/s41586-023-06936-2Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard.

《自然》（2023）。https://doi.org/10.1038/s41586-023-06936-2download引文接收日期：2022年7月6日接收日期：2023年12月5日发布日期：2023年12月12日OI：https://doi.org/10.1038/s41586-023-06936-2share本文与您共享以下链接的任何人都可以阅读此内容：获取可共享链接对不起，本文目前没有可共享的链接。复制到剪贴板。

Provided by the Springer Nature SharedIt content-sharing initiative

由Springer Nature SharedIt内容共享计划提供

CommentsBy submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.

评论通过提交评论，您同意遵守我们的条款和社区指南。如果您发现有虐待行为或不符合我们的条款或准则，请将其标记为不合适。