Decoding Human Biology and Disease with Single-Cell Omics Technologies

Background Introduction

Cells are the fundamental units of life. A single fertilized egg can develop into an entire complex human body, composed of approximately 37 trillion cells organized into various tissues, organs, and systems. Traditional cell classification methods primarily rely on cell morphology, location, or the expression levels of a few proteins, neglecting differences between cells at other molecular levels. The high heterogeneity of cells determines the functional diversity of human biology. Not only the cell’s own state, size, or origin but also the special environment surrounding the cell and interactions with adjacent or distant cells affect cellular characteristics. Traditional bulk sequencing techniques such as RNA sequencing obscure cellular diversity by averaging gene expression measurements across all cells in an experimental sample. Therefore, understanding human biology and disease at single-cell resolution is crucial.

Paper Source

This paper was written by researchers Qiang Shi, Xueyan Chen, and Zemin Zhang from the Biomedical Pioneering Innovation Center, School of Life Sciences, Peking University, and published in the journal “Genomics Proteomics & Bioinformatics” (2023) (DOI: 10.1016/j.gpb.2023.06.003). The paper mainly summarizes the latest developments in single-cell omics technologies and their applications in cancer research.

Development of Single-Cell Omics Technologies

The development of Single-Cell Omics (SCO) technologies aims to deconstruct cellular heterogeneity driven by intrinsic programs and extrinsic factors. All SCO technologies are designed to decode information surrounding DNA, RNA, and proteins, which are at the core of the central dogma of molecular biology. Here are the developments of SCO technologies at different molecular levels:

Genome

Single-cell whole genome sequencing has made significant progress in detecting genetic variations such as single nucleotide variations (SNVs), small insertions/deletions (Indels), copy number variations (CNVs), and structural variations (SVs). These variations, though occurring at low frequencies, gradually accumulate during the development, aging, and disease processes in multiple human tissue types. Emerging methods such as “Smooth-Seq” show superior performance in detecting structural variations (SVs) and extrachromosomal circular DNA (eccDNA).

Epigenetics

Epigenetic regulation (such as chromatin state, chromosome conformation, and DNA or histone modifications) plays a crucial role in the genetic networks of cells through the correlation between genetic information and its functional products. Measurement techniques for epigenetics at the single-cell level have been developed to explore chromatin accessibility, three-dimensional genome structure, and histone modification states.

Transcriptome

scRNA-seq technology has been widely applied, is cost-effective, and easy to master for non-professionals. Transcriptome sequencing excels in cell fate, cell state, and cell type classification, capable of detecting rare cell types and revealing the dynamic characteristics of cell fate transitions.

Multi-modal Omics

To more comprehensively characterize individual cells, single-cell multi-modal omics technologies have emerged, capable of simultaneously measuring multiple properties within a single cell. The application of CRISPR technology has made the study of genotype-phenotype associations more direct. These technologies have greatly accelerated our comprehensive understanding of gene variations, gene expression, intracellular regulatory networks, intercellular communication, and environmental effects.

Data Analysis

In terms of data analysis, the large amount of data generated by SCO technologies has driven the development of computational methods, forming a positive feedback loop. For example, the analysis pipeline for scRNA-seq involves data preprocessing, quality control, normalization, selection of highly variable genes (HVGs), dimensionality reduction, visualization, and automatic or manual annotation of cell types. Advanced analyses include differential expression and composition analysis, trajectory inference, gene regulatory network (GRNs) reconstruction, exploration of cell-cell interactions, and multi-modal integration. The construction of large-scale atlas datasets has also promoted the emergence of new annotation methods, such as the deep learning-based transfer learning method scBERT.

Research Applications

Human Complexity Research

scRNA-seq has been used to map the transcriptomic characteristics of various cell types and states, revealing new cell subgroups and functional features in studies of human blood, brain, heart, kidney, and other organs. Particularly in the field of cancer, SCO technologies have revealed the complexity of the tumor microenvironment (TME) and discovered potential immunotherapy targets.

Disease Analysis

SCO applications targeting diseases have revealed pathological features of many tissues. For example, single-cell analyses of scleroderma, Crohn’s disease, and lung cancer have revealed new pathological cell modules and gene regulatory features, providing new directions for disease diagnosis and treatment. During the COVID-19 pandemic, SCO technologies helped identify effective neutralizing antibodies, demonstrating their great potential in disease response.

Multi-modal Omics Research

Multi-modal omics integrates data from multiple molecular levels, providing a more comprehensive understanding of cellular heterogeneity and dynamic characteristics, such as simultaneous transcriptome and proteome sequencing, transcriptome and epigenome sequencing, etc. Although multi-modal omics technologies have made significant progress, single-cell multi-modal sequencing still faces many challenges, such as detection sensitivity and coverage issues. Additionally, studies combining spatial information and CRISPR perturbations will promote understanding of the complex relationships between genotype and phenotype.

Clinical Prospects

Single-cell omics technologies have broad clinical application prospects, including more precise patient classification and more personalized treatment methods. However, achieving their widespread clinical application requires overcoming bottlenecks such as high costs and technical complexity. Automating and simplifying data analysis workflows will be important steps in driving this technology into clinical practice.

Summary and Outlook

Single-cell omics technologies represent the enormous potential of precision medicine, providing powerful tools for human biology and disease research. In the future, with continuous technological advancements and expanding applications, SCO technologies will drive comprehensive systematic research from the cellular level to the tissue level, from genomics to epigenetics, revealing deeper mechanisms of human biology and disease. Ultimately, these technological advances will translate into more efficient medical services and better patient treatment outcomes.