CoChat
Sign Up
Cited by 38Single-cell and spatial transcriptomicsGene Regulatory Network AnalysisGene expression and cancer classification

MichiGAN: sampling from disentangled representations of single-cell data using generative adversarial networks

Open access
Yu, Hengshi, Welch, Joshua D.·2021·Genome Biology
DOI 10.1186/s13059-021-02373-4PMC PMC8139054

Published version of a preprint

Open this paper in CoChat — free accountOpen PDF

Abstract

Abstract Deep generative models such as variational autoencoders (VAEs) and generative adversarial networks (GANs) generate and manipulate high-dimensional images. We systematically assess the complementary strengths and weaknesses of these models on single-cell gene expression data. We also develop MichiGAN, a novel neural network that combines the strengths of VAEs and GANs to sample from disentangled representations without sacrificing data generation quality. We learn disentangled representations of three large single-cell RNA-seq datasets and use MichiGAN to sample from these representations. MichiGAN allows us to manipulate semantically distinct aspects of cellular identity and predict single-cell gene expression response to drug treatment.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1186/s13059-021-02373-4

Cites

6 of 6 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Yu et al. proposed MichiGAN, which combined VAEs and GANs to learn disentangled representations of single-cell datasets and generate single-cell datasets with biological insights.

    “Yu et al. proposed MichiGAN, which combined VAEs and GANs to learn the disentangled representations of single-cell datasets and generate single-cell datasets with biological insights [167].”

    From Challenges in AI-driven Biomedical Multimodal Data Fusion and Analysis (2025) · cites this paper as [167]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 72

  1. [1]LeCun Y Bengio Y Hinton G Deep learning Nature 2015 521 7553 436 44 10.1038/nature14539 26017442 · DOI · Paper page
  2. [2]Bengio Y Courville A Vincent P Representation learning: a review and new perspectives IEEE Trans Pattern Anal Mach Intell 2013 35 8 1798 828 10.1109/TPAMI.2013.50 23787338 · DOI · Paper page
  3. [3]Theis L, Oord A, Bethge M. A note on the evaluation of generative models. arXiv preprint arXiv:1511.01844. 2015. · arXiv
  4. [5]Wu Y, Donahue J, Balduzzi D, Simonyan K, Lillicrap T. Logan: Latent optimisation for generative adversarial networks. arXiv preprint arXiv:1912.00953. 2019. · arXiv
  5. [6]Kingma DP, Welling M. Auto-encoding variational Bayes. arXiv preprint arXiv:1312.6114. 2013. · arXiv
  6. [7]Goodfellow I, Pouget-Abadie J, Mirza M, Xu B, Warde-Farley D, Ozair S, Courville A, Bengio Y. Generative adversarial nets. In: Advances in Neural Information Processing Systems: 2014. p. 267280.
  7. [8]Goodfellow I, Bengio Y, Courville A. Deep Learning: MIT Press; 2016. http://www.deeplearningbook.org.
  8. [10]Larsen ABL, Snderby SK, Larochelle H, Winther O. Autoencoding beyond pixels using a learned similarity metric. In: International Conference on Machine Learning. PMLR: 2016. p. 155866.
  9. [11]Arjovsky M, Chintala S, Bottou L. Wasserstein gan. arXiv preprint arXiv:1701.07875. 2017. · arXiv
  10. [12]Gulrajani I, Ahmed F, Arjovsky M, Dumoulin V, Courville AC. Improved training of wasserstein gans. In: Advances in Neural Information Processing Systems: 2017. p. 576777.
  11. [13]Heusel M, Ramsauer H, Unterthiner T, Nessler B, Hochreiter S. Gans trained by a two time-scale update rule converge to a local nash equilibrium. In: Advances in Neural Information Processing Systems: 2017. p. 662637.
  12. [14]Desjardins G, Courville A, Bengio Y. Disentangling factors of variation via generative entangling. arXiv preprint arXiv:1210.5474. 2012. · arXiv
  13. [15]Ridgeway K. A survey of inductive biases for factorial representation-learning. arXiv preprint arXiv:1612.05299. 2016. · arXiv
  14. [16]Denton EL, et al. Unsupervised learning of disentangled representations from video. In: Advances in Neural Information Processing Systems: 2017. p. 441423.
  15. [18]Eastwood C, Williams CK. A framework for the quantitative evaluation of disentangled representations. In: International Conference on Learning Representations: 2018.
  16. [19]Locatello F, Bauer S, Lucic M, Raetsch G, Gelly S, Schlkopf B, Bachem O. Challenging common assumptions in the unsupervised learning of disentangled representations. In: International Conference on Machine Learning: 2019. p. 411424.
  17. [20]Higgins I, Amos D, Pfau D, Racaniere S, Matthey L, Rezende D, Lerchner A. Towards a definition of disentangled representations. arXiv preprint arXiv:1812.02230. 2018. · arXiv
  18. [21]Berthelot D, Raffel C, Roy A, Goodfellow I. Understanding and improving interpolation in autoencoders via an adversarial regularizer. arXiv preprint arXiv:1807.07543. 2018. · arXiv
  19. [22]Karras T, Laine S, Aila T. A style-based generator architecture for generative adversarial networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2019. p. 440110. 32012000
  20. [23]Hsu W-N, Zhang Y, Glass J. Unsupervised learning of disentangled and interpretable representations from sequential data. In: Advances in Neural Information Processing Systems: 2017. p. 187889.
  21. [24]Dupont E. Learning disentangled joint continuous and discrete representations. In: Advances in Neural Information Processing Systems: 2018. p. 71020.
  22. [25]Bai Y, Duan LL. Tuning-free disentanglement via projection. arXiv preprint arXiv:1906.11732. 2019. · arXiv
  23. [26]Rolinek M, Zietlow D, Martius G. Variational autoencoders pursue pca directions (by accident). In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2019. p. 1240615.
  24. [27]Esmaeili B, Wu H, Jain S, Bozkurt A, Siddharth N, Paige B, Brooks DH, Dy J, Meent J-W. Structured disentangled representations. In: The 22nd International Conference on Artificial Intelligence and Statistics. PMLR: 2019. p. 252534.
  25. [28]Khemakhem I, Kingma D, Monti R, Hyvarinen A. Variational autoencoders and nonlinear ica: A unifying framework. In: International Conference on Artificial Intelligence and Statistics: 2020. p. 220717.
  26. [29]Kim H, Mnih A. Disentangling by factorising. In: International Conference on Machine Learning: 2018. p. 264958.
  27. [30]Chen TQ, Li X, Grosse RB, Duvenaud DK. Isolating sources of disentanglement in variational autoencoders. In: Advances in Neural Information Processing Systems: 2018. p. 261020.
  28. [31]Gao S, Brekelmans R, Ver Steeg G, Galstyan A. Auto-encoding total correlation explanation. In: The 22nd International Conference on Artificial Intelligence and Statistics: 2019. p. 115766.
  29. [32]Chen X, Duan Y, Houthooft R, Schulman J, Sutskever I, Abbeel P. Infogan: Interpretable representation learning by information maximizing generative adversarial nets. In: Advances in Neural Information Processing Systems: 2016. p. 217280.
  30. [33]Ramesh A, Choi Y, LeCun Y. A spectral regularizer for unsupervised disentanglement. arXiv preprint arXiv:1812.01161. 2018. · arXiv
  31. [34]Kaneko T, Hiramatsu K, Kashino K. Generative adversarial image synthesis with decision tree latent controller. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2018. p. 660615.
  32. [36]Lin Z, Thekumparampil KK, Fanti G, Oh S. Infogan-cr: Disentangling generative adversarial networks with contrastive regularizers. arXiv preprint arXiv:1906.06034. 2019. · arXiv
  33. [37]Kazemi H, Iranmanesh SM, Nasrabadi N. Style and content disentanglement in generative adversarial networks. In: 2019 IEEE Winter Conference on Applications of Computer Vision (WACV). IEEE: 2019. p. 84856.
  34. [38]Shen Y, Gu J, Tang X, Zhou B. Interpreting the latent space of gans for semantic face editing. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition: 2020. p. 924352.
  35. [39]Liu B, Zhu Y, Fu Z, de Melo G, Elgammal A. Oogan: Disentangling gan with one-hot sampling and orthogonal regularization. In: AAAI: 2020. p. 483643.
  36. [40]Lee W, Kim D, Hong S, Lee H. High-fidelity synthesis with disentangled representation. In: European Conference on Computer Vision. Springer: 2020. p. 15774.
  37. [42]Paysan P, Knothe R, Amberg B, Romdhani S, Vetter T. A 3D face model for pose and illumination invariant face recognition. In: 2009 Sixth IEEE International Conference on Advanced Video and Signal Based Surveillance. IEEE: 2009. p. 296301.
  38. [43]Aubry M, Maturana D, Efros AA, Russell BC, Sivic J. Seeing 3d chairs: exemplar part-based 2d-3d alignment using a large dataset of cad models. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2014. p. 37629.
  39. [44]Liu Z, Luo P, Wang X, Tang X. Deep learning face attributes in the wild. In: Proceedings of the IEEE International Conference on Computer Vision: 2015. p. 37308.
  40. [45]Burgess CP, Higgins I, Pal A, Matthey L, Watters N, Desjardins G, Lerchner A. Understanding disentangling in -vae. arXiv preprint arXiv:1804.03599. 2018. · arXiv
  41. [46]White T. Sampling generative networks. arXiv preprint arXiv:1609.04468. 2016. · arXiv
  42. [48]Dosovitskiy A, Tobias Springenberg J, Brox T. Learning to generate chairs with convolutional neural networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition: 2015. p. 153846.
  43. [49]Sainburg T, Thielk M, Theilman B, Migliori B, Gentner T. Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions. arXiv preprint arXiv:1807.06650. 2018. · arXiv
  44. [51]Lotfollahi M Wolf FA Theis FJ scGen predicts single-cell perturbation responses Nat Methods 2019 16 8 715 21 10.1038/s41592-019-0494-8 31363220 · DOI · Paper page
  45. [52]Tan J, Ung M, Cheng C, Greene CS. Unsupervised feature construction and knowledge extraction from genome-wide assays of breast cancer with denoising autoencoders. In: Pacific Symposium on Biocomputing Co-Chairs. World Scientific: 2014. p. 13243. PMC4299935 25592575
  46. [53]Gupta A, Wang H, Ganapathiraju M. Learning structure in gene expression data using deep architectures, with an application to gene clustering. In: 2015 IEEE International Conference on Bioinformatics and Biomedicine (BIBM). sIEEE: 2015. p. 132835.
  47. [54]Way GP, Greene CS. Extracting a biologically relevant latent space from cancer transcriptomes with variational autoencoders. BioRxiv. 2017:174474. PMC5728678 29218871
  48. [55]Rampasek L, Hidru D, Smirnov P, Haibe-Kains B, Goldenberg A. Dr. vae: Drug response variational autoencoder. arXiv preprint arXiv:1706.08203. 2017. · arXiv
  49. [56]Deng Y, Bao F, Dai Q, Wu LF, Altschuler SJ. Massive single-cell rna-seq analysis and imputation via deep learning. bioRxiv. 2018:315556.
  50. [57]Grnbech CH, Vording MF, Timshel PN, Snderby CK, Pers TH, Winther O. scvae: Variational auto-encoders for single-cell gene expression datas. bioRxiv. 2018:318295. 32415966
  51. [58]Wang D Gu J VASC: dimension reduction and visualization of single-cell RNA-seq data by deep variational autoencoder Genomics Proteomics Bioinforma 2018 16 5 320 31 10.1016/j.gpb.2018.08.003 PMC6364131 30576740 · DOI
  52. [60]Hu Q, Greene CS. Parameter tuning is a key part of dimensionality reduction via deep variational autoencoders for single cell RNA transcriptomics. In: PSB. World Scientific: 2019. p. 36273. PMC6417816 30963075
  53. [61]Cui H Zhou C Dai X Liang Y Paffenroth R Korkin D Boosting gene expression clustering with system-wide biological information: a robust autoencoder approach Int J Comput Biol Drug Des 2020 13 1 98 123 10.1504/IJCBDD.2020.105113 · DOI
  54. [62]Marouf M Machart P Bansal V Kilian C Magruder DS Krebs CF Bonn S Realistic in silico generation and augmentation of single-cell RNA-seq data using generative adversarial networks Nat Commun 2020 11 1 1 12 10.1038/s41467-019-14018-z 31919373 PMC6952370 · DOI
  55. [63]Zappia L Phipson B Oshlack A Splatter: simulation of single-cell RNA sequencing data Genome Biol 2017 18 1 174 10.1186/s13059-017-1305-0 28899397 PMC5596896 · DOI · Paper page
  56. [64]Papadopoulos N Gonzalo PR Sding J PROSSTT: probabilistic simulation of single-cell RNA-seq data for complex differentiation processes Bioinformatics 2019 35 18 3517 9 10.1093/bioinformatics/btz078 30715210 PMC6748774 · DOI
  57. [66]Bastidas-Ponce A Tritschler S Dony L Scheibner K Tarquis-Medina M Salinno C Schirge S Burtscher I Bttcher A Theis FJ Lickert H Bakhti M Klein A Treutlein B Comprehensive single cell mRNA profiling reveals a detailed roadmap for pancreatic endocrinogenesis Development 2019 146 12 dev173849 10.1242/dev.173849 31160421 · DOI
  58. [68]Wolf FA Angerer P Theis FJ SCANPY: large-scale single-cell gene expression data analysis Genome Biol 2018 19 1 15 10.1186/s13059-017-1382-0 29409532 PMC5802054 · DOI · Paper page
  59. [69]Tipping ME Bishop CM Probabilistic principal component analysis J R Stat Soc Ser B Stat Methodol 1999 61 3 611 22 10.1111/1467-9868.00196 · DOI · Paper page
  60. [70]Barratt S, Sharma R. A note on the inception score. arXiv preprint arXiv:1801.01973. 2018. · arXiv
  61. [71]Pu Y, Wang W, Henao R, Chen L, Gan Z, Li C, Carin L. Adversarial symmetric variational autoencoder. In: Advances in Neural Information Processing Systems: 2017. p. 43309.
  62. [72]Mescheder L, Nowozin S, Geiger A. In: International Conference on Machine Learning. PMLR: 2017. p. 2391400.
  63. [73]Miyato T, Koyama M. cGANs with projection discriminator. arXiv preprint arXiv:1802.05637. 2018. · arXiv
  64. [74]Spurr A, Aksan E, Hilliges O. Guiding infogan with semi-supervision. In: Joint European Conference on Machine Learning and Knowledge Discovery in Databases. Springer: 2017. p. 11934.
  65. [76]Pei D Shu X Gassama-Diagne A Thiery JP Mesenchymalepithelial transition in development and reprogramming Nat Cell Biol 2019 21 1 44 53 10.1038/s41556-018-0195-z 30602762 · DOI
  66. [77]Mirza M, Osindero S. Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784. 2014. · arXiv
  67. [78]Reed S, Akata Z, Yan X, Logeswaran L, Schiele B, Lee H. Generative adversarial text to image synthesis. In: International Conference on Machine Learning. PMLR: 2016. p. 10609.
  68. [79]Odena A, Olah C, Shlens J. Conditional image synthesis with auxiliary classifier gans. In: Proceedings of the 34th International Conference on Machine Learning-Volume 70. JMLR. org: 2017. p. 264251.
  69. [80]Bishop CM Pattern Recognition and Machine Learning (Information Science and Statistics) 2006 Berlin Springer-Verlag
  70. [81]Yadav A, Shah S, Xu Z, Jacobs D, Goldstein T. Stabilizing adversarial nets with prediction methods. arXiv preprint arXiv:1705.07364. 2017. · arXiv
  71. [82]Yu H, Welch J. MichiGAN: sampling from disentangled representations of single-cell data using generative adversarial networks. Zenodo. 2021. 10.5281/zenodo.4728278. PMC8139054 34016135 · DOI
  72. [83]Yu H, Welch J. MichiGAN: sampling from disentangled representations of single-cell data using generative adversarial networks. Github. 2021. https://github.com/welch-lab/MichiGAN. PMC8139054 34016135

Cited by

Sorted by citations

Authors
Yu, H., & Welch, J. D.
Affiliations
University of Michigan (US)
Corresponding author
  • Yu, Hengshi
  • Welch, Joshua D.
ORCID
Venue
Genome Biology
Publisher
Springer Science and Business Media LLC
Published
2021-05-20
Volume
22
Issue
1
Pages
158-158
ISSN
  • 1465-6906 (ISSN-L)
  • 1474-760X
DOI
10.1186/s13059-021-02373-4
PMC
PMC8139054
OpenAlex
W3160682313
Cited by
38
References
6
Citations by year
2026: 4 · 2025: 12 · 2024: 6 · 2023: 8 · 2022: 7 · 2021: 1
FWCI
1.73
Type
journal article
Language
en
Open access
Licence
Locations
  • Genome biology (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • Deep Blue (University of Michigan) (submittedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
National Human Genome Research Institute (R01HG010883)
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Single-cell and spatial transcriptomics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene Regulatory Network Analysis
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Gene expression and cancer classification
Keywords
Michigan, disentangled representations, single-cell RNA sequencing, generative adversarial networks, variational autoencoder, generative models, data generation, single-cell gene expression, drug response prediction, cell identity
MeSH terms
RNA-Seq, Algorithms, Computer Simulation, Gene Expression Regulation, Humans, Neural Networks, Computer, Statistics, Nonparametric, Gene Expression Profiling, Single-Cell Analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review