Single-Cell Multiomics Integration by SCOT

Abstract

Although the availability of various sequencing technologies allows us to capture different genome properties at single-cell resolution, with the exception of a few co-assaying technologies, applying different sequencing assays on the same single cell is impossible. Single-cell alignment using optimal transport (SCOT) is an unsupervised algorithm that addresses this limitation by using optimal transport to align single-cell multiomics data. First, it preserves the local geometry by constructing a k-nearest neighbor (k-NN) graph for each data set (or domain) to capture the intra-domain distances. SCOT then finds a probabilistic coupling matrix that minimizes the discrepancy between the intra-domain distance matrices. Finally, it uses the coupling matrix to project one single-cell data set onto another through barycentric projection, thus aligning them. SCOT requires tuning only two hyperparameters and is robust to the choice of one. Furthermore, the Gromov-Wasserstein distance in the algorithm can guide SCOT's hyperparameter tuning in a fully unsupervised setting when no orthogonal alignment information is available. Thus, SCOT is a fast and accurate alignment method that provides a heuristic for hyperparameter selection in a real-world unsupervised single-cell data alignment scenario. We provide a tutorial for SCOT and make its source code publicly available on GitHub.

Keywords: data integration; manifold alignment; multiomics; optimal transport; single-cell genomics.

FIG. 1.

Overview of the SCOT algorithm: SCOT takes in two data sets of single-cell genomics measurements in the form of count matrices to align them. For each data set, it first constructs k-nearest neighbor graphs based on correlations between samples and calculates the distance matrices capturing the intra-domain distances. Given these, it solves the Gromov-Wasserstein optimal transport formulation to find an ideal coupling (also known as “correspondence”) matrix, describing the probability of alignment between the samples across the two data sets. Finally, it completes the alignment by projecting the first domain onto the second one based on correspondence probabilities using barycentric projection. SCOT, single-cell alignment using optimal transport.