Leveraging neighborhood representations of single-cell data to achieve sensitive DE testing with miloDE

Missarova, Alsu; Dann, Emma; Rosen, Leah; Satija, Rahul; Marioni, John

doi:10.1186/s13059-024-03334-3

Cited by 5 publications

(2 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For differential expression analysis across conditions, the state of the art is to take an integrated embedding, assign the cells to clusters, and find differentially expressed genes separately for each given cluster using methods known from bulk RNA-seq analysis ("pseudobulking") (Crowell et al, 2020;Missarova et al, 2024). Here, we turn this process around.…”

Section: Genesmentioning

confidence: 99%

See 1 more Smart Citation

Analysis of multi-condition single-cell data with latent embedding multivariate regression

Ahlmann-Eltze

Huber

2023

Preprint

View full text Add to dashboard Cite

Multi-condition single-cell data reveal expression differences between corresponding cell subpopulations in different conditions. Current approaches divide cells into discrete groups or clusters and identify differentially expressed genes between corresponding groups. Here, we propose a method that operates without such grouping. Latent embedding multivariate regression (LEMUR) is based on a parametric mapping of latent space representations into each other and uses a design matrix to encode categorical and continuous covariates. We use the method to analyze a drug treatment experiment on brain tumor biopsies. We detect drug-induced gene expression responses affecting subsets of cells in a continuous latent space representation that does not require discrete categorization of the cells. Latent embedding multivariate regression is a versatile new approach for identifying differentially expressed genes from single-cell data of heterogeneous cell subpopulations or tissues under arbitrary experimental or study designs.

show abstract

Section: Genesmentioning

confidence: 99%

“…3G upper panel). In addition, it was more powerful than a pseudobulked test across all cells (global ), or separate tests for subsets of cells, either by cell type or cluster as in Crowell et al (2020), or by neighborhood as in miloDE (Missarova et al, 2024) (Fig. 3G, lower panel).…”

Section: Performance Assessmentmentioning

confidence: 99%

Analysis of multi-condition single-cell data with latent embedding multivariate regression

Ahlmann-Eltze

Huber

2023

Preprint

View full text Add to dashboard Cite

show abstract

Tumor stage-driven disruption of NK cell maturation in human and murine tumors

Russick,

Torset,

Sun

et al. 2024

iScience

View full text Add to dashboard Cite

Estimating the impact of single-cell RNA sequencing of human tissues on drug target validation

Dann,

Teeple,

Elmentaite

et al. 2024

Preprint

View full text Add to dashboard Cite

Early characterization of drug targets associated with disease can greatly reduce clinical failures attributed to lack of safety or efficacy. As single-cell RNA sequencing (scRNA-seq) of human tissues becomes increasingly common for disease profiling, the insights obtained from this data could influence target selection strategies. Whilst the use of scRNA-seq to understand target biology is well established, the impact of single-cell data in increasing the probability of candidate therapeutic targets to successfully advance from research to clinic has not been fully characterized. Inspired by previous work on an association between genetic evidence and clinical success, we used retrospective analysis of known drug target genes to identify potential predictors of target clinical success from scRNA-seq data. Particularly, we investigated whether successful drug targets are associated with cell type specific expression in a disease-relevant tissue (cell type specificity) or cell type specific over-expression in disease patients compared to healthy controls (disease cell specificity). Analysing scRNA-seq data across 30 diseases and 13 tissues, we found that both classes of scRNA-seq support significantly increase the odds of clinical success for gene-disease pairs. We estimate that combined they could approximately triple the chances of a target reaching phase III. Importantly, scRNA-seq analysis identifies a larger and complementary target space to that of direct genetic evidence. In particular, scRNA-seq support is more likely to prioritize therapeutically tractable classes of genes such as membrane-bound proteins. Our study suggests that scRNA-seq-derived information on cell type- and disease-specific expression can be leveraged to identify tractable and disease-relevant targets, with increased probability of success in the clinic.

show abstract

Leveraging neighborhood representations of single-cell data to achieve sensitive DE testing with miloDE

Cited by 5 publications

References 87 publications

Analysis of multi-condition single-cell data with latent embedding multivariate regression

Analysis of multi-condition single-cell data with latent embedding multivariate regression

Tumor stage-driven disruption of NK cell maturation in human and murine tumors

Estimating the impact of single-cell RNA sequencing of human tissues on drug target validation

Contact Info

Product

Resources

About