Single-cell Multi-omics Integration for Unpaired Data by a Siamese Network with Graph-based Contrastive Loss

Abstract

Single-cell omics technology is being rapidly developed to measure the epigenome, genome, and transcriptome across a range of cell types. However, integrating omics data from different modalities is still challenging. Here, we propose a variation of the Siamese neural network framework called MinNet, which is trained to integrate multi-omics data on the single-cell resolution by utilizing graph-based contrastive loss. By training the model and testing it on several benchmark datasets, we showed its accuracy and generalizability in integrating scRNA-seq with scATAC-seq, and scRNA-seq with epitopes data. Further evaluation demonstrated our models unique capacity in removing the batch effect, which is a common problem in actual practice. To show how the integration impacts downstream analysis, we established model-based smoothing and cis-regulatory element inferring method and validated it with external pcHi-C evidence. Finally, the framework was applied to a COVID-19 dataset to compensate the original work with integration-based analysis, showing its necessity in single-cell multi-omics research.