Tuning parameters of dimensionality reduction methods for single-cell RNA-seq analysis

Raimundo, Félix; Vallot, Céline; Vert, Jean-Philippe

doi:10.1101/2020.04.27.064816

Cited by 8 publications

(14 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This assumption is however challenged by recent studies which highlight significant differences in gene expression patterns 11,16 , mostly resulting from the lack of micro-environment in cell lines ( Figure 1 A). To assess this cell-type drift and evaluate the ability of existing batch-effect correction tools to correct it, we employed three state-of-the-art approaches 33,34 : Seurat v3 29 , Harmony 35 and LIGER 32 (Methods). For tumors, we use a panel of 208,506 cells from 58 NSCLC cancer patients at different disease stages 20 , including 36,467 epithelial cells ( Extended Figure 1 A-B), referred to as the Kim dataset .…”

Section: Resultsmentioning

confidence: 99%

Identifying commonalities between cell lines and tumors at the single cell level using Sobolev Alignment of deep generative models

Mourragui

Siefert

Reinders

et al. 2022

Preprint

View full text Add to dashboard Cite

Preclinical models are essential to cancer research, however, key biological differences with patient tumors result in reduced translatability to the clinic and high attrition rates in drug development. Variability among and between patients, preclinical models, and individual cells obscures commonalities which could otherwise be exploited therapeutically. To discover the shared biological processes between cell line models and clinical tumors we developed Sobolev Alignment, a computational framework which uses deep generative models to capture non-linear processes in single-cell RNA sequencing data and kernel methods to align and interpret these processes. We show that our approach faithfully captures shared processes on a set of three synthetic datasets. Exploiting two large panels of untreated non-small cell lung cancer cell lines and patients, we identify the similarities between cell lines and tumors and show the conservation of key mitotic and immune-related pathways. Employing our approach on a large in-vitro perturbation screen, we show that processes captured by our method faithfully recapitulate the known modes of action of clinically approved drugs and allow investigation into the mode of action of an uncharacterized drug.

show abstract

Section: Resultsmentioning

confidence: 99%

Identifying commonalities between cell lines and tumors at the single cell level using Sobolev Alignment of deep generative models

Mourragui

Siefert

Reinders

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…We removed cells three median absolute deviations (MAD) under the mean in counts and expressed genes, as well as those three MAD above the mean in percentage of mitochondrial reads. The code and data used in this manuscript, as well as the 5,000 pre-computed embeddings provided as a benchmark to develop new heuristics for parameter selection, are freely available under an Apache License 2.0 [30,31].…”

Section: Methodsmentioning

confidence: 99%

Tuning parameters of dimensionality reduction methods for single-cell RNA-seq analysis

Raimundo¹,

Vallot

Vert³

2020

Genome Biol

Self Cite

View full text Add to dashboard Cite

Background: Many computational methods have been developed recently to analyze single-cell RNA-seq (scRNA-seq) data. Several benchmark studies have compared these methods on their ability for dimensionality reduction, clustering, or differential analysis, often relying on default parameters. Yet, given the biological diversity of scRNA-seq datasets, parameter tuning might be essential for the optimal usage of methods, and determining how to tune parameters remains an unmet need. Results: Here, we propose a benchmark to assess the performance of five methods, systematically varying their tunable parameters, for dimension reduction of scRNA-seq data, a common first step to many downstream applications such as cell type identification or trajectory inference. We run a total of 1.5 million experiments to assess the influence of parameter changes on the performance of each method, and propose two strategies to automatically tune parameters for methods that need it. Conclusions: We find that principal component analysis (PCA)-based methods like scran and Seurat are competitive with default parameters but do not benefit much from parameter tuning, while more complex models like ZinbWave, DCA, and scVI can reach better performance but after parameter tuning.

show abstract

“…Approaches like scVI can generally yield an improvement over shallow approaches such as PCA (Raimundo et al 2020). Although there is not always a benefit in using deep models over shallow approaches Bellot et al (2018), in theory, deep approaches should always result in a better performance given enough data being available.…”

Section: Training Data and Hyperparameter Optimizationmentioning

confidence: 99%

Interpretable generative deep learning: an illustration with single cell gene expression data

2022

View full text Add to dashboard Cite

Deep generative models can learn the underlying structure, such as pathways or gene programs, from omics data. We provide an introduction as well as an overview of such techniques, specifically illustrating their use with single-cell gene expression data. For example, the low dimensional latent representations offered by various approaches, such as variational auto-encoders, are useful to get a better understanding of the relations between observed gene expressions and experimental factors or phenotypes. Furthermore, by providing a generative model for the latent and observed variables, deep generative models can generate synthetic observations, which allow us to assess the uncertainty in the learned representations. While deep generative models are useful to learn the structure of high-dimensional omics data by efficiently capturing non-linear dependencies between genes, they are sometimes difficult to interpret due to their neural network building blocks. More precisely, to understand the relationship between learned latent variables and observed variables, e.g., gene transcript abundances and external phenotypes, is difficult. Therefore, we also illustrate current approaches that allow us to infer the relationship between learned latent variables and observed variables as well as external phenotypes. Thereby, we render deep learning approaches more interpretable. In an application with single-cell gene expression data, we demonstrate the utility of the discussed methods.

show abstract

Tuning parameters of dimensionality reduction methods for single-cell RNA-seq analysis

Cited by 8 publications

References 29 publications

Identifying commonalities between cell lines and tumors at the single cell level using Sobolev Alignment of deep generative models

Identifying commonalities between cell lines and tumors at the single cell level using Sobolev Alignment of deep generative models

Tuning parameters of dimensionality reduction methods for single-cell RNA-seq analysis

Interpretable generative deep learning: an illustration with single cell gene expression data

Contact Info

Product

Resources

About