SIMBA: SIngle-cell eMBedding Along with features

Abstract

Recent advances in single cell omics technologies enable the individual or joint profiling of cellular measurements including gene expression, epigenetic features, chromatin structure and DNA sequences. Currently, most single-cell analysis pipelines are cluster-centric, i.e., they first cluster cells into non-overlapping cellular states and then extract their defining genomic features. These approaches assume that discrete clusters correspond to biologically relevant subpopulations and do not explicitly model the interactions between different feature types. However, cellular processes are defined in individual cells and inherently involve multiple genomic features that interact with each other and together provide complementary views on principles of gene regulation. In addition, single-cell methods are generally designed for a particular task as distinct single-cell problems are formulated differently. To address these current shortcomings, we present SIMBA, a single-cell embedding method that embeds single cells along with their defining features, such as genes, chromatin accessible regions, and transcription factor binding sequences, into a common latent space. By leveraging the co-embedding of cells and features, SIMBA allows for cellular heterogeneity study, clustering-free marker discovery, gene regulation inference, batch effect removal, and omics data integration. SIMBA has been extensively applied to scRNA-seq, scATAC-seq, and dual-omics data. We show that SIMBA provides a single framework that allows diverse single-cell analysis problems to be formulated in a common way and thus simplifies the development of new analyses and integration of other single-cell modalities.

Quantitatively visualizing bipartite datasets

“…As datasets continue to grow in size and complexity, it becomes increasingly important to quantitatively visualize interactions between entities. Future datasets may require multi-localization, where higher-order interactions (e.g., between a ligand and multimeric receptor (22); antibodies, antigens, and cell receptors (52); or single-cell multi-omics datasets (53)) are embedded in a low-dimensional space.…”

A CRISPR/Cas9-based enhancement of high-throughput single-cell transcriptomics

“…SIMBA (1.1) (https://github.com/pinellolab/simba) was used to embed individual genes alongside cells in feature space, enabling the characterization of cell type-gene specificity 82 . Cellranger output files filtered_feature_bc_matrix.h5 for each condition were input and consequently subset to only the cells that remained post filtering and were used for downstream analysis (see PBMC Secondary Analysis).…”

Scalable sequence-informed embedding of single-cell ATAC-seq data with CellSpace

“…However, scBasset requires training of a large neural network model where the number of tasks equals the number of cells and likely will require further optimizations to scale to large datasets. Finally, a recent method called SIMBA uses a graph embedding approach for scRNA-seq, scATAC-seq, and multiome data 6 , where cells, genes, peaks, k -mers, and TF motifs are vertices in the graph for a data set, and edges connect entities (like peaks) that relate to other entities (like cells). Notably for the application of this method to scATAC-seq, the TF motifs must be specified prior to training in order to define the graph and therefore will impact the learned embedding.…”