Reading the heart at single-cell resolution
Introduction
Cells are the most basic units of life. The ability to obtain transcriptomic landscapes of single cells has transformed our view of the human body, which spurred the Human Cell Atlas Project [1,2]. Started in 2017, this project is an international effort of hundreds of scientists aiming to classify and map all of the estimated 37 trillion cells in the human body. Single-cell RNA sequencing (scRNA-seq) and complementary techniques, such as spatial transcriptomics, are capable of reconstructing organs at ultra-high resolution in both temporal and spatial dimensions. A spatiotemporal cell atlas of the developing human heart has recently been reported [3,4], showcasing part of the fruitful achievements of the continuous efforts to comprehend the biology of the heart at single-cell resolution.
Studies involving scRNA-seq are inherently interdisciplinary as biology and bioinformatics are both integral components. The latter has rapidly evolved over the past decade, with many software packages developed each year to facilitate the detection of real and meaningful biological information and minimize the impacts of technical artifacts. Numerous factors are known to distort data output, including, but not restricted to, minute amounts of starting material leading to biases and dropouts, cellular integrity, batch effects, curse of dimensionality, etc. [[5], [6], [7], [8]]. Therefore, high quality biological samples and sequencing data are necessary for reliable single-cell analysis. In the following text, we summarize key points with respect to ensuring the quality of both sample and data in cardiac studies (Fig. 1). Once high-quality sequencing data is produced, aim-directed bioinformatic analyses coupled with rigorous wet-lab validation and careful interpretation comprise the final steps in reaching informative conclusions. Thus, in addition to outlining popular applications of scRNA-seq, we provide examples of context-dependent analyses (Table 1) for gaining biological insights.
Section snippets
Sample preparation
Single-cell preparation is the critical first step of all single-cell studies. The heart consists of multiple cell types, including cardiomyocytes, the muscle cells of the heart, and other supporting cell types such as endothelial cells, fibroblasts, and immune cells, which are collectively called non-cardiomyocytes or non-myocytes. Since methods for isolating non-myocytes and embryonic/fetal, as well as neonatal, cardiomyocytes are already established [[9], [10], [11]], and even commercially
Technical variability
To interrogate biological phenomena, it is vital to accurately estimate and then account for technical variability (Fig. 2). The sources of biologically irrelevant variation are collectively termed “batch effects”, and have been under intense scrutiny since the advent of high-throughput sequencing [43,44]. Technical variations contribute significantly to the overall variability in single-cell transcriptome analysis, and if not accounted for, can significantly confound data interpretation and
Bioinformatic analyses and data interpretation
The ultimate goal of scRNA-seq is to look through the mist of sequencing reads, and discern real biological phenomena. In concert with complex computations, scRNA-seq enables the assessment of cellular heterogeneity, identification of rare cell types, elucidation of cellular dynamics, and unraveling of intercellular crosstalk. These and other applications have radically changed our perception of the heart. Here, we elaborate on three fundamental types of bioinformatic analysis that are
Future perspectives
In the era of molecular cell typing, single-cell RNA sequencing has penetrated nearly all areas of biological and biomedical research. Fortified with increasingly versatile sequencing platforms, as well as newer and faster algorithms, scRNA-seq has virtually become indispensible in portraying the vast cellular heterogeneity and complex dynamics of organ systems. Unscrambling data can nevertheless be hard, and researchers are advised to strictly adhere to protocols for maximizing sample quality
Disclosures
The authors report no commercial or proprietary interest in any product or concept discussed in this article.
Acknowledgments
This work is supported by the National Key R&D Program of China (2017YFA0103700), CAMS Innovation Fund for Medical Sciences (CIFMS, 2018-I2M-3-002, 2017-I2M-1-003), Non-profit Central Research Institute Fund of Chinese Academy of Medical Sciences (2019PT320026), and grants (31671542, 81722006 to L.W., 81700337 to B.Z.) from the National Natural Science Foundation of China.
References (127)
- et al.
Single-Cell resolution of temporal gene expression during heart development
Dev. Cell
(2016) - et al.
Transcriptomic profiling maps anatomically patterned subpopulations among single embryonic cardiac cells
Dev. Cell
(2016) - et al.
High-throughput single-molecule RNA imaging analysis reveals heterogeneous responses of cardiomyocytes to hemodynamic overload
J. Mol. Cell. Cardiol.
(2019) - et al.
Single-cell imaging and transcriptomic analyses of endogenous cardiomyocyte dedifferentiation and cycling
Cell Discov
(2019) - et al.
Nuclear retention of mRNA in mammalian tissues
Cell Rep.
(2015) - et al.
Dynamics of Cell generation and turnover in the Human heart
Cell
(2015) - et al.
The technology and biology of single-cell RNA sequencing
Mol. Cell
(2015) - et al.
An introduction to the analysis of single-Cell RNA-sequencing data
Mol Ther Methods Clin Dev
(2018) - et al.
The human mitochondrial transcriptome
Cell
(2011) - et al.
Why batch effects matter in Omics data, and how to avoid them
Trends Biotechnol.
(2017)