Constructing Cell-type Taxonomy by Optimal Transport with Relaxed Marginal Constraints

TL;DR

This paper introduces a novel method combining Optimal Transport with Relaxed Marginal Constraints to construct a comprehensive cell-type taxonomy across multiple samples, improving annotation accuracy and handling unmatched or unique cell clusters.

Contribution

It presents a new system that effectively aligns and annotates cell clusters across multiple datasets, accommodating varying cluster proportions and unique cell types.

Findings

Achieves high accuracy in cell type annotation across twenty datasets.

Enables effective alignment of clusters with varying proportions and unmatched types.

Facilitates accurate sample classification based on extracted features.

Abstract

The rapid emergence of single-cell data has facilitated the study of many different biological conditions at the cellular level. Cluster analysis has been widely applied to identify cell types, capturing the essential patterns of the original data in a much more concise form. One challenge in the cluster analysis of cells is matching clusters extracted from datasets of different origins or conditions. Many existing algorithms cannot recognize new cell types present in only one of the two samples when establishing a correspondence between clusters obtained from two samples. Additionally, when there are more than two samples, it is advantageous to align clusters across all samples simultaneously rather than performing pairwise alignment. Our approach aims to construct a taxonomy for cell clusters across all samples to better annotate these clusters and effectively extract features for downstream analysis. A new system for constructing cell-type taxonomy has been developed by combining the technique of Optimal Transport with Relaxed Marginal Constraints (OT-RMC) and the simultaneous alignment of clusters across multiple samples. OT-RMC allows us to address challenges that arise when the proportions of clusters vary substantially between samples or when some clusters do not appear in all the samples. Experiments on more than twenty datasets demonstrate that the taxonomy constructed by this new system can yield highly accurate annotation of cell types. Additionally, sample-level features extracted based on the taxonomy result in accurate classification of samples.