Infusing Structural Assumptions into Dimensionality Reduction for Single-cell RNA Sequencing Data to Identify Small Gene Sets

Overview

Journal Commun Biol

Specialty Biology

Date 2025 Mar 12

PMID 40069486

Authors

Maren Hackenberg

Niklas Brunn

Tanja Vogel

Harald Binder

Affiliations

Soon will be listed here.

Abstract

Dimensionality reduction greatly facilitates the exploration of cellular heterogeneity in single-cell RNA sequencing data. While most of such approaches are data-driven, it can be useful to incorporate biologically plausible assumptions about the underlying structure or the experimental design. We propose the boosting autoencoder (BAE) approach, which combines the advantages of unsupervised deep learning for dimensionality reduction and boosting for formalizing assumptions. Specifically, our approach selects small sets of genes that explain latent dimensions. As illustrative applications, we explore the diversity of neural cell identities and temporal patterns of embryonic development.

References

Binder H, Schumacher M . Allowing for mandatory covariates in boosting estimation of sparse high-dimensional survival models. BMC Bioinformatics. 2008; 9:14. PMC: 2245904. DOI: 10.1186/1471-2105-9-14. View

Tasic B, Menon V, Nguyen T, Kim T, Jarsky T, Yao Z . Adult mouse cortical cell taxonomy revealed by single cell transcriptomics. Nat Neurosci. 2016; 19(2):335-46. PMC: 4985242. DOI: 10.1038/nn.4216. View

Treppner M, Binder H, Hess M . Interpretable generative deep learning: an illustration with single cell gene expression data. Hum Genet. 2022; 141(9):1481-1498. PMC: 9360114. DOI: 10.1007/s00439-021-02417-6. View

Lu C, Dong L, Zhou H, Li Q, Huang G, Bai S . G-Protein-Coupled Receptor Gpr17 Regulates Oligodendrocyte Differentiation in Response to Lysolecithin-Induced Demyelination. Sci Rep. 2018; 8(1):4502. PMC: 5852120. DOI: 10.1038/s41598-018-22452-0. View

Ashuach T, Gabitto M, Koodli R, Saldi G, Jordan M, Yosef N . MultiVI: deep generative model for the integration of multimodal data. Nat Methods. 2023; 20(8):1222-1231. PMC: 10406609. DOI: 10.1038/s41592-023-01909-9. View

Luecken M, Buttner M, Chaichoompu K, Danese A, Interlandi M, Mueller M . Benchmarking atlas-level data integration in single-cell genomics. Nat Methods. 2021; 19(1):41-50. PMC: 8748196. DOI: 10.1038/s41592-021-01336-8. View

Hess M, Hackenberg M, Binder H . Exploring generative deep learning for omics data using log-linear models. Bioinformatics. 2020; 36(20):5045-5053. PMC: 7755415. DOI: 10.1093/bioinformatics/btaa623. View

Jindal A, Gupta P, Jayadeva , Sengupta D . Discovery of rare cells from voluminous single cell expression data. Nat Commun. 2018; 9(1):4719. PMC: 6226447. DOI: 10.1038/s41467-018-07234-6. View

Schlosser P, Knaus J, Schmutz M, Dohner K, Plass C, Bullinger L . Netboost: Boosting-Supported Network Analysis Improves High-Dimensional Omics Prediction in Acute Myeloid Leukemia and Huntington's Disease. IEEE/ACM Trans Comput Biol Bioinform. 2020; 18(6):2635-2648. DOI: 10.1109/TCBB.2020.2983010. View

10.

Gronbech C, Vording M, Timshel P, Sonderby C, Pers T, Winther O . scVAE: variational auto-encoders for single-cell gene expression data. Bioinformatics. 2020; 36(16):4415-4422. DOI: 10.1093/bioinformatics/btaa293. View

11.

Tutz G, Binder H . Generalized additive modeling with implicit variable selection by likelihood-based boosting. Biometrics. 2006; 62(4):961-71. DOI: 10.1111/j.1541-0420.2006.00578.x. View

12.

Eraslan G, Simon L, Mircea M, Mueller N, Theis F . Single-cell RNA-seq denoising using a deep count autoencoder. Nat Commun. 2019; 10(1):390. PMC: 6344535. DOI: 10.1038/s41467-018-07931-2. View

13.

Choi Y, Li R, Quon G . siVAE: interpretable deep generative models for single-cell transcriptomes. Genome Biol. 2023; 24(1):29. PMC: 9940350. DOI: 10.1186/s13059-023-02850-y. View

14.

Lotfollahi M, Susmelj A, De Donno C, Hetzel L, Ji Y, Ibarra I . Predicting cellular responses to complex perturbations in high-throughput screens. Mol Syst Biol. 2023; 19(6):e11517. PMC: 10258562. DOI: 10.15252/msb.202211517. View

15.

Saelens W, Cannoodt R, Todorov H, Saeys Y . A comparison of single-cell trajectory inference methods. Nat Biotechnol. 2019; 37(5):547-554. DOI: 10.1038/s41587-019-0071-9. View

16.

Azodi C, Tang J, Shiu S . Opening the Black Box: Interpretable Machine Learning for Geneticists. Trends Genet. 2020; 36(6):442-455. DOI: 10.1016/j.tig.2020.03.005. View

17.

Buettner F, Pratanwanich N, McCarthy D, Marioni J, Stegle O . f-scLVM: scalable and versatile factor analysis for single-cell RNA-seq. Genome Biol. 2017; 18(1):212. PMC: 5674756. DOI: 10.1186/s13059-017-1334-8. View

18.

Huang H, Wang Y, Rudin C, Browne E . Towards a comprehensive evaluation of dimension reduction methods for transcriptomic data visualization. Commun Biol. 2022; 5(1):719. PMC: 9296444. DOI: 10.1038/s42003-022-03628-x. View

19.

Moon K, van Dijk D, Wang Z, Gigante S, Burkhardt D, Chen W . Visualizing structure and transitions in high-dimensional biological data. Nat Biotechnol. 2019; 37(12):1482-1492. PMC: 7073148. DOI: 10.1038/s41587-019-0336-3. View

20.

Cakir B, Prete M, Huang N, van Dongen S, Pir P, Kiselev V . Comparison of visualization tools for single-cell RNAseq data. NAR Genom Bioinform. 2020; 2(3):lqaa052. PMC: 7391988. DOI: 10.1093/nargab/lqaa052. View