SCIM: Universal Single-Cell Matching with Unpaired Feature Sets

Abstract

Motivation: Recent technological advances have led to an increase in the production and availability of single-cell data. The ability to integrate a set of multi-technology measurements would allow the identification of biologically or clinically meaningful observations through the unification of the perspectives afforded by each technology. In most cases, however, profiling technologies consume the used cells and thus pairwise correspondences between datasets are lost. Due to the sheer size single-cell datasets can acquire, scalable algorithms that are able to universally match single-cell measurements carried out in one cell to its corresponding sibling in another technology are needed. Results: We propose Single-Cell data Integration via Matching (SCIM), a scalable approach to recover such correspondences in two or more technologies. SCIM assumes that cells share a common (low-dimensional) underlying structure and that the underlying cell distribution is approximately constant across technologies. It constructs a technology-invariant latent space using an auto-encoder framework with an adversarial objective. Multi-modal datasets are integrated by pairing cells across technologies using a bipartite matching scheme that operates on the low-dimensional latent representations. We evaluate SCIM on a simulated cellular branching process and show that the cell-to-cell matches derived by SCIM reflect the same pseudotime on the simulated dataset. Moreover, we apply our method to two real-world scenarios, a melanoma tumor sample and a human bone marrow sample, where we pair cells from a scRNA dataset to their sibling cells in a CyTOF dataset achieving 93% and 84% cell-matching accuracy for each one of the samples respectively.

Inferring and perturbing cell fate regulomes in human cerebral organoids

“…For each time point and line separately, we performed Canonical Correlation Analysis (CCA) on gene activities and gene expression data using the Seurat function based on 2000 features, which were selected using the Seurat function . In CCA space, we performed minimum-cost maximum-flow (MCMF) bipartite matching between the modalities as described (20) (https://github.com/ratschlab/scim). The function was used with , , and otherwise default parameters.…”

“…S1, B to D). We constructed 'multi-omic metacells' containing information on both transcriptome and chromatin accessibility using minimum-cost, maximum-flow (MCMF) bipartite matching (20) within the CCA space (Fig. 1A).…”

pmVAE: Learning Interpretable Single-Cell Representations with Pathway Modules

“…While features sets might not be easily comparable, the pathway structure underlying these feature sets are shared. This indicates that pathway-factorized latent representations, like those learned by pmVAE, could be used more easily integrate [5,9,28,43] these technologies.…”

Multi-omics integration and regulatory inference for unpaired single-cell data with a graph-linked unified embedding framework

“…Apart from the feature-converted data, Seurat v3 15 and bindSC 30 also devised heuristic strategies to utilize information in the original feature space, which probably explains their improved performance than methods that do not 16, 17 . At the cell level, known cell types have also been used via (semi-)supervised learning 47, 48 , but this approach incurs substantial limitations in terms of applicability since such supervision is typically unavailable and in many cases serves as the purpose of multi-omics integration per se 26 . Notably, one of these methods was proposed with a similar autoencoder architecture and adversarial alignment 48 , but it relied on matched cell types or clusters to orient the alignment.…”