Single-cell RNA-seq data semi-supervised clustering and annotation via structural regularized domain adaptation

Abstract

Motivation The rapid development of single-cell RNA sequencing (scRNA-seq) technologies allows us to explore tissue heterogeneity at the cellular level. The identification of cell types plays an essential role in the analysis of scRNA-seq data, which, in turn, influences the discovery of regulatory genes that induce heterogeneity. As the scale of sequencing data increases, the classical method of combining clustering and differential expression analysis to annotate cells becomes more costly in terms of both labor and resources. Existing scRNA-seq supervised classification method can alleviate this issue through learning a classifier trained on the labeled reference data and then making a prediction based on the unlabeled target data. However, such label transference strategy carries with risks, such as susceptibility to batch effect and further compromise of inherent discrimination of target data. Results In this paper, inspired by unsupervised domain adaptation, we propose a flexible single cell semi-supervised clustering and annotation framework, scSemiCluster, which integrates the reference data and target data for training. We utilize structure similarity regularization on the reference domain to restrict the clustering solutions of the target domain. We also incorporates pairwise constraints in the feature learning process such that cells belonging to the same cluster are close to each other, and cells belonging to different clusters are far from each other in the latent space. Notably, without explicit domain alignment and batch effect correction, scSemiCluster outperforms other state-of-the-art, single-cell supervised classification and semi-supervised clustering annotation algorithms in both simulation and real data. To the best of our knowledge, we are the first to use both deep discriminative clustering and deep generative clustering techniques in the single-cell field. Availability An implementation of scSemiCluster is available from https://github.com/xuebaliang/scSemiCluster. Supplementary information Supplementary notes are available at Bioinformatics online.

Clustering CITE-seq data with a canonical correlation-based deep learning method

“…First, we transform and to low-dimensional representations and with omics-specific encoder networks f ( r ) and f ( p ) separately. To capture the character of transcriptomic (scRNA-seq) data, we utilize the same ZINB (zero-inflated negative binomial) ( Huang et al, 2018 ) model-based autoencoder as in scSemiCluster ( Chen et al, 2021 ). Protein data are not as sparse as scRNA-seq data; therefore, we empirically use a negative binomial (NB) model, a ZINB model hybrid, to characterize it.…”

Cell-type Annotation with Accurate Unseen Cell-type Identification Using Multiple References

“…In order to enable the embedding component E ij to fully capture the characteristics of cells and make the classification model better fit the query dataset, we employ the embedding component as an encoder and use a mirror image of the embedding component as a decoder to construct an autoencoder [16, 17]. The reconstruction loss of cells both from the reference subset and the query subset is taken into consideration when training the classification model.…”

A Brief Review on Deep Learning Applications in Genomic Studies

“…Both methods use variational inference and deep generative models to fully characterize the distribution of single-cell RNA-seq data. scANVI can also be used to annotate cell types and has been used as a baseline method for recently proposed methods, such as scSemiCluster (Chen et al, 2021) and scNym (Kimmel and Kelley, 2021). scReClassify proposed by Kim et al (2019) uses PCA to perform dimension reduction of the original single cell RNAseq data, and then apply a semi-supervised learning method to reclassify the mislabeled cell types caused by human inspection.…”