Contrastive latent variable modeling with application to case-control sequencing experiments

Jones, Andrew; Townes, F. William; Li, Didong; Engelhardt, Barbara E.

doi:10.48550/arxiv.2102.06731

Cited by 5 publications

(17 citation statements)

References 41 publications

(67 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Isolating salient variations present only in a target dataset is the subject of contrastive analysis (CA) [56,3,22,29,40,2,48]. While many recent studies have modeled scRNAseq data by fitting probabilistic models and representing the data in a lower dimension [30,38,18,32,31], few of these models are designed for CA.…”

Section: Mainmentioning

confidence: 99%

“…scVI has achieved state-of-the-art results on many tasks; however, it was not specifically designed for the CA setting and thus may struggle to capture salient variations in target samples. We also compared against two contrastive methods designed for analyzing scRNA-seq count data: contrastive Poisson latent variable model (CPLVM) and contrastive generalized latent variable model (CGLVM) [22]. While these methods are designed for the contrastive setting, they both make the strong assumption that linear models can accurately capture the complex variations in scRNA-seq data.…”

Section: Contrastivevi Outperforms Other Modeling Approachesmentioning

confidence: 99%

“…Previous studies have successfully leveraged Perturb-Seq to better understand regulatory circuits related to innate immunity [21], the unfolded protein response pathway [4], and the T cell receptor signaling pathway [10], among other applications. Despite these successes, recent work [22] has suggested that naive approaches for analyzing Perturb-Seq data may fail to capture subtle perturbation-induced transcriptomic changes due to the presence of intercellular variations unrelated to the perturbations. Thus, methods for isolating variations unique to the perturbed cells may unlock new biological insights missed by previous approaches.…”

Section: Contrastivevi Stratifies Crispr-perturbed Cells By Induced G...mentioning

confidence: 99%

“…For example, data from healthy controls versus a diseased population or from pre-versus post-intervention groups form intuitive background and target pairs. Moreover, with the development of new technologies for measuring cellular responses to large numbers of perturbations in parallel, such as Perturb-Seq [11], MIX-Seq [35], MULTI-Seq [36], and sci-Plex [41] among others, tools for better understanding variations unique to such perturbed cells compared to control populations will be critical.Isolating salient variations present only in a target dataset is the subject of contrastive analysis (CA) [56,3,22,29,40,2,48]. While many recent studies have modeled scRNAseq data by fitting probabilistic models and representing the data in a lower dimension [30,38,18,32,31], few of these models are designed for CA.…”

mentioning

confidence: 99%

“…Such methods are thus unlikely to capture the enriched variations in a target dataset, which are often subtle compared to the overall variations in the data [3]. One recent study [22] designed a probabilistic model for analyzing scRNA-seq data in the CA setting. However, this method assumes that a linear model is sufficiently expressive to model the variations in scRNA-seq data, even though previous work has demonstrated substantial improvements by using more expressive nonlinear methods [30].To address these limitations, we developed contrastiveVI, a deep generative model that enables analysis of scRNA-seq data in the CA setting.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Isolating salient variations of interest in single-cell data with contrastiveVI

Weinberger

Lin

Lee

2021

Preprint

View full text Add to dashboard Cite

Single-cell RNA sequencing (scRNA-seq) technologies enable a better understanding of previously unexplored biological diversity. Oftentimes, researchers are specifically interested in modeling the latent structures and variations enriched in one target scRNA-seq dataset as compared to another background dataset generated from sources of variation irrelevant to the task at hand. For example, we may wish to isolate factors of variation only present in measurements from patients with a given disease as opposed to those shared with data from healthy control subjects. Here we introduce Contrastive Variational Inference (contrastiveVI; https://github.com/suinleelab/contrastiveVI), a framework for end-to-end analysis of target scRNA-seq datasets that decomposes the variations into shared and target-specific factors of variation. On three target-background dataset pairs we demonstrate that contrastiveVI learns latent representations that recover known subgroups of target data points better than previous methods and finds differentially expressed genes that agree with known ground truths.

show abstract

Section: Mainmentioning

confidence: 99%

Section: Contrastivevi Outperforms Other Modeling Approachesmentioning

confidence: 99%

Section: Contrastivevi Stratifies Crispr-perturbed Cells By Induced G...mentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 3 more Smart Citations

Isolating salient variations of interest in single-cell data with contrastiveVI

Weinberger

Lin

Lee

2021

Preprint

View full text Add to dashboard Cite

show abstract

A Poisson reduced-rank regression model for association mapping in sequencing data

Fitzgerald

Jones

Engelhardt

2022

Preprint

Self Cite

View full text Add to dashboard Cite

Single-cell RNA sequencing (scRNA-seq) technologies allow for the study of gene expression in individual cells. Often, it is of interest to understand how transcriptional activity is associated with cell-specific covariates, such as cell type, genotype, or measures of cell health. Traditional approaches for this type of association mapping assume independence between the outcome variables (or genes), and perform a separate regression for each. However, these methods are computationally costly and ignore the substantial correlation structure of gene expression. Furthermore, count-based scRNA-seq data pose challenges for traditional models based on Gaussian assumptions. We aim to resolve these issues by developing a reduced-rank regression model that identifies low-dimensional linear associations between a large number of cell-specific covariates and high-dimensional gene expression readouts. Our probabilistic model uses a Poisson likelihood in order to account for the unique structure of scRNA-seq counts. We demonstrate the performance of our model using simulations, and we apply our model to a scRNA-seq dataset, a spatial gene expression dataset, and a bulk RNA-seq dataset to show its behavior in three distinct analyses. We show that our statistical modeling approach, which is based on reduced-rank regression, captures associations between gene expression and cell- and sample-specific covariates by leveraging low-dimensional representations of transcriptional states.

show abstract

BuDDI:Bulk Deconvolution with Domain Invarianceto predict cell-type-specific perturbations from bulk

Davidson¹,

Greene²

2023

Preprint

View full text Add to dashboard Cite

While single-cell experiments provide deep cellular resolution within a single sample, some single-cell experiments are inherently more challenging than bulk experiments due to dissociation difficulties, cost, or limited tissue availability. This creates a situation where we have deep cellular profiles of one sample or condition, and bulk profiles across multiple samples and conditions. To bridge this gap, we propose BuDDI (BUlk Deconvolution with Domain Invariance). BuDDI utilizes domain adaptation techniques to effectively integrate available corpora of case-control bulk and reference scRNA-seq observations to infer cell-type-specific perturbation effects. BuDDI achieves this by learning independent latent spaces within a single variational autoencoder (VAE) encompassing at least four sources of variability: 1) cell-type proportion, 2) perturbation effect, 3) structured experimental variability, and 4) remaining variability. Since each latent space is encouraged to be independent, we simulate perturbation responses by independently composing each latent space to simulate cell-type-specific perturbation responses. We evaluated BuDDI's performance on simulated and real data with experimental designs of increasing complexity. We first validated that BuDDI could learn domain invariant latent spaces on data with matched samples across each source of variability. Then we validated that BuDDI could accurately predict cell-type-specific perturbation response when no single-cell perturbed profiles were used during training; instead, only bulk samples had both perturbed and non-perturbed observations. Finally, we validated BuDDI on predicting sex-specific differences, an experimental design where it is not possible to have matched samples. In each experiment, BuDDI outperformed all other comparative methods and baselines. As more reference atlases are completed, BuDDI provides a path to combine these resources with bulk-profiled treatment or disease signatures to study perturbations, sex differences, or other factors at single-cell resolution.

show abstract

Contrastive latent variable modeling with application to case-control sequencing experiments

Cited by 5 publications

References 41 publications

Isolating salient variations of interest in single-cell data with contrastiveVI

Isolating salient variations of interest in single-cell data with contrastiveVI

A Poisson reduced-rank regression model for association mapping in sequencing data

BuDDI:Bulk Deconvolution with Domain Invarianceto predict cell-type-specific perturbations from bulk

Contact Info

Product

Resources

About