Clustering single-cell RNA-seq data by rank constrained similarity learning

Abstract

Motivation Recent breakthroughs of single-cell RNA sequencing (scRNA-seq) technologies offer an exciting opportunity to identify heterogeneous cell types in complex tissues. However, the unavoidable biological noise and technical artifacts in scRNA-seq data as well as the high dimensionality of expression vectors make the problem highly challenging. Consequently, although numerous tools have been developed, their accuracy remains to be improved. Results Here, we introduce a novel clustering algorithm and tool RCSL (Rank Constrained Similarity Learning) to accurately identify various cell types using scRNA-seq data from a complex tissue. RCSL considers both local similarity and global similarity among the cells to discern the subtle differences among cells of the same type as well as larger differences among cells of different types. RCSL uses Spearman’s rank correlations of a cell’s expression vector with those of other cells to measure its global similarity, and adaptively learns neighbour representation of a cell as its local similarity. The overall similarity of a cell to other cells is a linear combination of its global similarity and local similarity. RCSL automatically estimates the number of cell types defined in the similarity matrix, and identifies them by constructing a block-diagonal matrix, such that its distance to the similarity matrix is minimized. Each block-diagonal submatrix is a cell cluster/type, corresponding to a connected component in the cognate similarity graph. When tested on 16 benchmark scRNA-seq datasets in which the cell types are well-annotated, RCSL substantially outperformed six state-of-the-art methods in accuracy and robustness as measured by three metrics. Availability The RCSL algorithm is implemented in R and can be freely downloaded at https://cran.r-project.org/web/packages/RCSL/index.html. Supplementary information Supplementary data are available at Bioinformatics online.

Robust joint clustering of multi-omics single-cell data via multi-modal high-order neighborhood Laplacian matrix optimization

“…When data are in large scale, efficient hierarchical clustering algorithm was developed ( Zou et al 2021 ). Also, method focused on similarity learning was proposed in Mei et al (2021) for identifying cell types using scRNA-seq data. Recent progress has shed light on graph attention auto-encoder for scRNA-seq data representation and clustering ( Cheng and Ma 2022 ).…”

Algorithms for a Commons Cell Atlas

“…First, instead of joining multiple batches together to perform assignment, we instead perform assignment on a per matrix basis (we term each matrix an “observation”, Methods). Second, instead of performing assignments on the UMI count matrix, mx assign performs assignments on matrices normalized using ranks (Vargo and Gilbert 2020; Franzén, Gan, and Björkegren 2019; Mei, Li, and Su 2021). This means the distance measurement via Euclidean distance in the GMM is replaced with the Spearman correlation (Methods).…”

scMAE: a masked autoencoder for single-cell RNA-seq clustering

“…SIMLR ( Wang et al 2017 ) combines multiple kernel functions with a multi-kernel Bayesian learning algorithm during the dimensionality reduction process to capture different features and relationships in the data. RCSL ( Mei et al 2021 ) constructs a similarity matrix by measuring both global and local relationships between cells and then derives a block diagonal matrix from it to obtain the final clustering results. SMSC ( Qi et al 2021 ) adopts a multiple kernel combination approach, enabling direct learning of similarity metrics from single-cell RNA sequencing data while simultaneously considering the constraints of clustering structure, thereby discovering effective cell clusters.…”