A joint deep learning model enables simultaneous batch effect correction, denoising, and clustering in single-cell transcriptomics

Abstract

Recent developments of single-cell RNA-seq (scRNA-seq) technologies have led to enormous biological discoveries. As the scale of scRNA-seq studies increases, a major challenge in analysis is batch effects, which are inevitable in studies involving human tissues. Most existing methods remove batch effects in a low-dimensional embedding space. Although useful for clustering, batch effects are still present in the gene expression space, leaving downstream gene-level analysis susceptible to batch effects. Recent studies have shown that batch effect correction in the gene expression space is much harder than in the embedding space. Methods such as Seurat 3.0 rely on the mutual nearest neighbor (MNN) approach to remove batch effects in gene expression, but MNN can only analyze two batches at a time, and it becomes computationally infeasible when the number of batches is large. Here, we present CarDEC, a joint deep learning model that simultaneously clusters and denoises scRNA-seq data while correcting batch effects both in the embedding and the gene expression space. Comprehensive evaluations spanning different species and tissues showed that CarDEC outperforms Scanorama, DCA + Combat, scVI, and MNN. With CarDEC denoising, non-highly variable genes offer as much signal for clustering as the highly variable genes (HVGs), suggesting that CarDEC substantially boosted information content in scRNA-seq. We also showed that trajectory analysis using CarDEC's denoised and batch-corrected expression as input revealed marker genes and transcription factors that are otherwise obscured in the presence of batch effects. CarDEC is computationally fast, making it a desirable tool for large-scale scRNA-seq studies.

Integrating genetic regulation and single-cell expression with GWAS prioritizes causal genes and cell types for glaucoma

“…The gene expression was then normalized by total UMI counts for each cell, scaled by 10000, and transformed with the natural logarithm. To account for batch effects originating from sample collection processes, the CarDEC framework 88 was applied with standard parameters. Both highly variable genes (top 3000) and lowly variable genes were used for building the model, and donor resources were used as the batch key.…”

Pathway Centric Analysis for single-cell RNA-seq and Spatial Transcriptomics Data with GSDensity

“…Imputation of unmeasured genes could be a strategy to overcome this limitation. Also, for integrated datasets from multiple origins with strong batch effects, since the MCA uses the original gene expression data, the batch effect correction should also be made on the original gene expression data, instead of the low-dimensional embeddings, using methods such as combat 46 , CarDEC 47 , or Scanorama 47 , etc. Our original motivation of developing GSDensity is to build a tool for 'pathway-centric' analysis, with which we can dive into the data directly from the angle of pathways of interest.…”

Fast model-free standardization and integration of single-cell transcriptomics data

“…There are many well-established integration methods available to address batch effects in scRNA-seq datasets, for example Seurat, Harmony, and LIGER 16,52,53 . Additionally, some deep learning-based methods such as HDMC and CarDEC are also available 54,55 . In this study, we rigorously tested cell-type score-based integration via MASI across various single-cell platforms, cytoplasm/nuclei, research groups, conditions, and individuals.…”