Single-cell assignment using multiple-adversarial domain adaptation network with large-scale references

The rapid accumulation of single-cell RNA-seq data has provided rich resources to characterize various hu-man cell populations. However, achieving accurate cell-type annotation using public references presents challenges due to inconsistent annotations, batch effects, and rare cell types. Here, we introduce SELINA (single-cell identity navigator), an integrative and automatic cell-type annotation framework based on a pre-curated reference atlas spanning various tissues. SELINA employs a multiple-adversarial domain adap-tation network to remove batch effects within the reference dataset. Additionally, it enhances the annotation of less frequent cell types by synthetic minority oversampling and fits query data with the reference data us-ing an autoencoder. SELINA culminates in the creation of a comprehensive and uniform reference atlas, en-compassing 1.7 million cells covering 230 distinct human cell types. We substantiate its robustness and su-periority across a multitude of human tissues. Notably, SELINA could accurately annotate cells within diverse disease contexts. SELINA provides a complete solution for human single-cell RNA-seq data annotation with both python and R packages.

Automated summary

SELINA is a comprehensive and automatic cell-type annotation framework based on a pre-curated reference atlas. It employs a multiple-adversarial domain adaptation network to remove batch effects and enhances the annotation of less frequent cell types through synthetic minority oversampling and autoencoder fitting. SELINA creates a comprehensive and uniform reference atlas with 1.7 million cells covering 230 distinct human cell types, and accurately annotates cells within diverse human tissues.