Multi-Omics of Single Cells: Strategies and Applications

Abstract

Most genome-wide assays provide averages across large numbers of cells, but recent technological advances promise to overcome this limitation. Pioneering single-cell assays are now available for genome, epigenome, transcriptome, proteome, and metabolome profiling. Here, we describe how these different dimensions can be combined into multi-omics assays that provide comprehensive profiles of the same cell.

Keywords: bioinformatic methods; cell state profiling; combined genome/epigenome/transcriptome/proteome/metabolome mapping; molecular medicine; single-cell analysis; single-cell systems biology.

Figure 1

Strategies for Multi-Omics Profiling of Single Cells. Conceptual diagram (top) and examples (bottom) showing five complementary strategies for measuring two different omics dimensions (represented by horizontal and vertical lines) in the same cell. The ‘Combine’ approach measures both dimensions in the same experiment (example: protein and metabolite profiles measured by mass spectrometry). The ‘Separate’ approach enriches two types of biomolecule in different fractions and analyzes them in separate experiments (example: DNA and RNA separated with beads). The ‘Split’ approach uses a fraction of the total cell lysate for each experiment (example: RNA and protein analyzed based on different fractions). The ‘Convert’ approach transforms one omics dimension to another and then analyzes the latter (example: DNA methylation and DNA sequence). The computational ‘Predict’ approach measures one omics dimension directly and bioinformatically infers the second based on the data for the first (example: DNA methylation and transcription factor occupancy). Collectively, these approaches provide building blocks that can be adapted and combined to design protocols for integrated analysis of the genome, epigenome, transcriptome, proteome, and/or metabolome of single cells.

Figure I

Schematic of Multi-Omics Data Analysis and Integration.