Partial domain adaptation enables cross domain cell type annotation between scRNA-seq and snRNA-seq

TL;DR

Experiments show that ScNucAdapt achieves robust and accurate cell type annotation, outperforming existing approaches, and provides a practical framework for the cross-domain cell type annotation between scRNA-seq and snRNA seq data.

Abstract

Accurate cell type annotation across datasets is a key challenge in single-cell analysis. snRNA-seq enables profiling of frozen or difficult-to-dissociate tissues, complementing scRNA-seq by capturing fragile or rare cell types. However, cross-annotation between these two datasets remains largely unexplored, as existing methods treat them independently. We introduce ScNucAdapt, a method designed for cross-annotation between paired and unpaired scRNA-seq and snRNA-seq datasets. To address distributional and cell composition differences, ScNucAdapt employs partial domain adaptation. Experiments across both unpaired and paired scRNA-seq and snRNA-seq show that ScNucAdapt achieves robust and accurate cell type annotation, outperforming existing approaches. Therefore, ScNucAdapt provides a practical framework for the cross-domain cell type annotation between scRNA-seq and snRNA seq data.

Figures (8)

• Figure 1: Concept of pair and unpaired scRNA-seq and snRNA-seq
• Figure 2: Concept of partial domain adaptation
• Figure 3: Overall framework of ScNucAdapt
• Figure 4: Simulation Experiments using Splatter (A) Accuracy on imbalanced and balanced simulated datasets domain shift experiments (B) Macro f1-score on imbalanced and balanced simulated datasets domain shift experiments (C) Accuracy on imbalanced and balanced simulated datasets class variation experiments (D) Macro f1-score on imbalanced and balanced simulated datasets class variation experiments (E) Uncorrected simulated dataset of cell types colored (F) cell type representations before source classes and target clusters merging (G) cell type representations after source classes and target clusters merging (H) Uncorrected simulated dataset of batch colored (I) batch representations before source classes and target clusters merging (J) batch representations after source classes and target clusters merging.
• Figure 5: Visualization result of scRNA-seq and snRNA-seq representations. Using UMAP on kidney tissue (A) visualization result on scRNA-seq to snRNA batch representations (B) visualization result on scRNA-seq to snRNA cell type representations