• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利：CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法，用中文像聊天一样搜索，搜遍4000万医学文献。AI智能推荐，让科研检索更轻松。

立即免费搜索

文件翻译

保留排版，准确专业，支持PDF/Word/PPT等文件格式，支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述，25分钟生成高质量综述，智能提取关键信息，辅助科研写作。

立即免费体验

scPreGAN，一种用于预测单细胞表达对扰动响应的深度生成模型。

scPreGAN, a deep generative model for predicting the response of single-cell expression to perturbation.

机构信息

Shanghai Key Lab of Intelligent Information Processing, Shanghai, China.

School of Computer Science and Technology, Fudan University, Shanghai, China.

出版信息

Bioinformatics. 2022 Jun 27;38(13):3377-3384. doi: 10.1093/bioinformatics/btac357.

DOI:10.1093/bioinformatics/btac357
PMID:35639705
Abstract

MOTIVATION

Rapid developments of single-cell RNA sequencing technologies allow study of responses to external perturbations at individual cell level. However, in many cases, it is hard to collect the perturbed cells, such as knowing the response of a cell type to the drug before actual medication to a patient. Prediction in silicon could alleviate the problem and save cost. Although several tools have been developed, their prediction accuracy leaves much room for improvement.

RESULTS

In this article, we propose scPreGAN (Single-Cell data Prediction base on GAN), a deep generative model for predicting the response of single-cell expression to perturbation. ScPreGAN integrates autoencoder and generative adversarial network, the former is to extract common information of the unperturbed data and the perturbed data, the latter is to predict the perturbed data. Experiments on three real datasets show that scPreGAN outperforms three state-of-the-art methods, which can capture the complicated distribution of cell expression and generate the prediction data with the same expression abundance as the real data.

AVAILABILITY AND IMPLEMENTATION

The implementation of scPreGAN is available via https://github.com/JaneJiayiDong/scPreGAN. To reproduce the results of this article, please visit https://github.com/JaneJiayiDong/scPreGAN-reproducibility.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

摘要

动机

单细胞 RNA 测序技术的快速发展使得能够在单细胞水平上研究对外界扰动的反应。然而，在许多情况下，很难收集受扰细胞，例如在给患者实际用药之前，就知道某种细胞类型对药物的反应。硅基预测可以缓解这个问题并节省成本。尽管已经开发了几种工具，但它们的预测准确性仍有很大的改进空间。

结果

在本文中，我们提出了 scPreGAN（基于 GAN 的单细胞数据预测），这是一种用于预测单细胞表达对扰动反应的深度生成模型。scPreGAN 集成了自动编码器和生成对抗网络，前者用于提取未受扰数据和受扰数据的共同信息，后者用于预测受扰数据。在三个真实数据集上的实验表明，scPreGAN 优于三种最先进的方法，它可以捕捉到细胞表达的复杂分布，并生成与真实数据具有相同表达丰度的预测数据。

可用性和实现

scPreGAN 的实现可通过 https://github.com/JaneJiayiDong/scPreGAN 获取。要重现本文的结果，请访问 https://github.com/JaneJiayiDong/scPreGAN-reproducibility。

补充信息

补充数据可在生物信息学在线获得。

相似文献

1
scPreGAN, a deep generative model for predicting the response of single-cell expression to perturbation.scPreGAN，一种用于预测单细胞表达对扰动响应的深度生成模型。
Bioinformatics. 2022 Jun 27;38(13):3377-3384. doi: 10.1093/bioinformatics/btac357.
2
scSemiGAN: a single-cell semi-supervised annotation and dimensionality reduction framework based on generative adversarial network.scSemiGAN：基于生成对抗网络的单细胞半监督注释和降维框架。
Bioinformatics. 2022 Nov 15;38(22):5042-5048. doi: 10.1093/bioinformatics/btac652.
3
Predicting single-cell cellular responses to perturbations using cycle consistency learning.使用循环一致性学习预测单细胞对扰动的细胞反应。
Bioinformatics. 2024 Jun 28;40(Suppl 1):i462-i470. doi: 10.1093/bioinformatics/btae248.
4
AI4AVP: an antiviral peptides predictor in deep learning approach with generative adversarial network data augmentation.AI4AVP：一种采用生成对抗网络数据增强的深度学习方法的抗病毒肽预测器。
Bioinform Adv. 2022 Oct 26;2(1):vbac080. doi: 10.1093/bioadv/vbac080. eCollection 2022.
5
HDMC: a novel deep learning-based framework for removing batch effects in single-cell RNA-seq data.HDMC：一种用于去除单细胞 RNA-seq 数据中批次效应的新型深度学习框架。
Bioinformatics. 2022 Feb 7;38(5):1295-1303. doi: 10.1093/bioinformatics/btab821.
6
scBGEDA: deep single-cell clustering analysis via a dual denoising autoencoder with bipartite graph ensemble clustering.scBGEDA：基于双分图集成分聚类的对偶去噪自动编码器的单细胞聚类分析。
Bioinformatics. 2023 Feb 14;39(2). doi: 10.1093/bioinformatics/btad075.
7
Single-cell RNA-seq data semi-supervised clustering and annotation via structural regularized domain adaptation.基于结构正则化领域自适应的单细胞 RNA-seq 数据半监督聚类和注释。
Bioinformatics. 2021 May 5;37(6):775-784. doi: 10.1093/bioinformatics/btaa908.
8
ResPAN: a powerful batch correction model for scRNA-seq data through residual adversarial networks.ResPAN：通过残差对抗网络对 scRNA-seq 数据进行强大的批量校正模型。
Bioinformatics. 2022 Aug 10;38(16):3942-3949. doi: 10.1093/bioinformatics/btac427.
9
Dr.VAE: improving drug response prediction via modeling of drug perturbation effects.VAE 博士：通过建模药物干扰效应来改善药物反应预测。
Bioinformatics. 2019 Oct 1;35(19):3743-3751. doi: 10.1093/bioinformatics/btz158.
10
Protein contact map refinement for improving structure prediction using generative adversarial networks.利用生成对抗网络进行蛋白质接触图优化以改进结构预测
Bioinformatics. 2021 Oct 11;37(19):3168-3174. doi: 10.1093/bioinformatics/btab220.

引用本文的文献

1
Protocol for predicting single- and multiple-dose-dependent gene expression using deep generative learning.使用深度生成学习预测单剂量和多剂量依赖性基因表达的方案。
STAR Protoc. 2025 Jul 11;6(3):103932. doi: 10.1016/j.xpro.2025.103932.
2
CoupleVAE: coupled variational autoencoders for predicting perturbational single-cell RNA sequencing data.CoupleVAE：用于预测扰动单细胞RNA测序数据的耦合变分自编码器。
Brief Bioinform. 2025 Mar 4;26(2). doi: 10.1093/bib/bbaf126.
3
scVAEDer: integrating deep diffusion models and variational autoencoders for single-cell transcriptomics analysis.
scVAEDer：整合深度扩散模型和变分自编码器用于单细胞转录组学分析
Genome Biol. 2025 Mar 21;26(1):64. doi: 10.1186/s13059-025-03519-4.
4
GEMDiff: a diffusion workflow bridges between normal and tumor gene expression states: a breast cancer case study.GEMDiff：一种连接正常与肿瘤基因表达状态的扩散工作流程：乳腺癌案例研究
Brief Bioinform. 2025 Mar 4;26(2). doi: 10.1093/bib/bbaf093.
5
Fatecode enables cell fate regulator prediction using classification-supervised autoencoder perturbation.命运编码通过分类监督自编码器扰动实现细胞命运调控因子预测。
Cell Rep Methods. 2024 Jul 15;4(7):100819. doi: 10.1016/j.crmeth.2024.100819. Epub 2024 Jul 9.
6
scPRAM accurately predicts single-cell gene expression perturbation response based on attention mechanism.scPRAM 基于注意力机制准确预测单细胞基因表达扰动响应。
Bioinformatics. 2024 May 2;40(5). doi: 10.1093/bioinformatics/btae265.
7
scButterfly: a versatile single-cell cross-modality translation method via dual-aligned variational autoencoders.scButterfly：一种通过双对齐变分自动编码器实现的多功能单细胞跨模态转换方法。
Nat Commun. 2024 Apr 6;15(1):2973. doi: 10.1038/s41467-024-47418-x.
8
scDisInFact: disentangled learning for integration and prediction of multi-batch multi-condition single-cell RNA-sequencing data.scDisInFact：用于多批次多条件单细胞 RNA-seq 数据的集成和预测的解缠学习。
Nat Commun. 2024 Jan 30;15(1):912. doi: 10.1038/s41467-024-45227-w.
9
Homogeneous Space Construction and Projection for Single-Cell Expression Prediction Based on Deep Learning.基于深度学习的单细胞表达预测的齐性空间构建与投影
Bioengineering (Basel). 2023 Aug 23;10(9):996. doi: 10.3390/bioengineering10090996.
10
Generative modeling of single-cell gene expression for dose-dependent chemical perturbations.用于剂量依赖性化学扰动的单细胞基因表达生成建模。
Patterns (N Y). 2023 Aug 11;4(8):100817. doi: 10.1016/j.patter.2023.100817.