Sci Signal. Nat Methods. 2015;161:1187201. Some studies also provided tantalizing glimpses of new biological phenomena that could not have been easily observed without scRNA-seq. Journal of Neuroinflammation 2020 Aug;40(8):329-344. doi: 10.1002/cac2.12078. 2015;31:298998. Before further analyses, scRNA-seq data typically require a number of bio-informatic QC checks, where poor-quality data from single cells (arising as a result of many possible reasons, including poor cell viability at the time of lysis, poor mRNA recovery and low efficiency of cDNA production) can be justifiably excluded from subsequent analysis. 2013;14:R31. RNA sequencing; gene expression profiling; single-cell analysis. Sequencing is a technology that allows us to read the sequence of DNA or RNA. Kiselev VY, Kirschner K, Schaub MT, Andrews T, Yiu A, Chandra T, et al. Microbiology and Molecular Biology Reviews: MMBR, 68, 538-559. doi: 10.1128/MMBR.68.3.538-559.2004. Hagemann-Jensen, M. et al. Wagner GP, Kin K, Lynch VJ. Martin Hemberg. The data suggest that, if the main goal of the study is to characterize the transcriptome of a particular cell with the greatest possible resolution, then a median read depth of around one million is essential. Visualizing Single-Cell RNA-Seq Data with t-SNE: Researcher Interview with Dmitry Kobak and Philipp Berens. Sci Rep. 2016;6:33892. 2012;36:14252. PubMed Central Cell. Barcoding Tagging single cells or sequencing libraries with unique oligonucleotide sequences (that is, barcodes), allowing sample multiplexing. Gierahn TM, Wadsworth MH, Hughes TK, Bryson BD, Butler A, Satija R, et al. 2009;6:37782. [32] and Svensson et al. Reconstructing lineage hierarchies of the distal lung epithelium using single-cell RNA-seq. RNA-Seq of single prostate CTCs implicates noncanonical Wnt signaling in antiandrogen resistance. Single-cell RNA-seq transcriptome analysis of linear and circular RNAs in mouse preimplantation embryos. Lopez, R., Regier, J., Cole, M. B., Jordan, M. I. Riemondy KA, Ransom M, Alderman C, Gillen AE, Fu R, Finlay-Schultz J, Kirkpatrick GD, Di Paola J, Kabos P, Sartorius CA, Hesselberth JR. Nucleic Acids Res. 2016;17:106. Lun, A. T. L., Bach, K. & Marioni, J. C. Pooling across cells to normalize single-cell RNA sequencing data with many zero counts. Science 352, 189196 (2016). Cell Rep. 2014;7:113042. Here we present an overview of the computational workflow involved in processing scRNA-seq data. Open Access We expect that many new combination approaches will emerge using proteomics, epigenomics and analysis of non-coding RNA species alongside scRNA-seq (reviewed in [100]). Nature. 28, 100108 (1979). volume9, Articlenumber:75 (2017) In summary, generating a robust scRNA-seq dataset is now feasible for wet-lab researchers with little to no prior expertise in single-cell genomics. Once the scRNA-seq data are filtered for poor samples, they can be interpreted by an ever-increasing range of bio-informatic and computational methods, which have been reviewed extensively elsewhere [74, 82]. An increasing number of algorithms and computational approaches are being published to help researchers define the molecular relationships between single cells characterized by scRNA-seq and thus extend the insights gained by simple clustering. 2017;18:45. Science. MeSH Sheng K, Cao W, Niu Y, Deng Q, Zong C. Effective detection of variation in single-cell transcriptomes using MATQ-seq. Power analysis of single-cell RNA-sequencing experiments. Jiang P, Thomson JA, Stewart R. Quality control of single-cell RNA-seq by SinQC. 30, 195204 (2020). Yet, while scRNA-seq can provide answers to many research questions, it is important to understand that the details of any answers provided will vary according to the protocol used. The data from spike-ins can be used for assessing the level of technical variability and for identifying genes with a high degree of biological variability [7]. Cell. Since its introduction, single-cell RNA sequencing (scRNA-seq) approaches have revolutionized the genomics field as they created unprecedented opportunities for resolving cell heterogeneity by exploring gene expression profiles at a single-cell resolution. 2011;21:11607. Street, K. et al. Moreover, owing to the digital nature of gene expression at the single-cell level, and the related phenomenon of transcriptional bursting (in which pulses of transcriptional activity are followed by inactive refractory periods; Box 1), transcript levels are subject to temporal fluctuation, further contributing to the high frequency of zero observations in scRNA-seq data. While the required number of cells is dependent on the number of distinct cell states within the population, the required sequencing depth also depends on the magnitude of differences between these states. Nat. Bioinformatics 36, 11741181 (2020). Here, we provide you with a brief explanation. Medicine now exists in a cellular and molecular era, where experimental biologists and clinicians seek to understand and modify cell behaviour through targeted molecular approaches. 2017. doi:10.1002/1873-3468.12684. 2014;343:1936. Cell. Single cell RNA-sequencing of pluripotent states unlocks modular transcriptional variation. 2017;14:26770. Diaz A, Liu SJ, Sandoval C, Pollen A, Nowakowski TJ, Lim DA, et al. Image credit: courtesy of Dr. Ayshwarya . Mol Biol Cell. Genome Biol. Once single cells have been deposited into individual wells of a plate, these protocols, and others from additional commercial suppliers (for example, BD Life Sciences/Cellular Research), can be conducted without the need for further expensive hardware other than accurate multi-channel pipettes, although it should be noted that, in the absence of a microfluidic platform in which to perform scRNA-seq reactions (for example, the C1 platform from Fluidigm), reaction volumes and therefore reagent costs can increase substantially. Single mammalian cells compensate for differences in cellular volume and DNA copy number through independent global transcriptional mechanisms. SC3: consensus clustering of single-cell RNA-seq data. scRNA-seq is also increasingly being used to trace lineage and developmental relationships between heterogeneous, yet related, cellular states in scenarios such as embryonal development, cancer, myoblast and lung epithelium differentiation and lymphocyte fate diversification [11,22,23,24,, 2125]. Reinius B, Mold JE, Ramskld D, Deng Q, Johnsson P, Michalsson J, et al. PubMed Lacar B, Linker SB, Jaeger BN, Krishnaswami S, Barron J, Kelder M, et al. Genome Biol. Haghverdi L, Buettner F, Theis FJ. Transcriptional bursting A phenomenon, also known as transcriptional pulsing, of relatively short transcriptionally active periods being followed by longer silent periods, resulting in temporal fluctuation of transcript levels. Nat. 10, 4667 (2019). Unsupervised clustering cannot integrate prior knowledge where relevant . We speculate that the next decade will take us closer to a truly holistic examination of single cells, which takes into account not only mRNA, but also the genome, epigenome, proteome and metabolome. The frequency of dropout events for scRNA-seq is protocol-dependent, and is closely associated with the number of sequencing reads generated for each cell (Svensson et al., 2017). After quality control and doublet removal, 13,744 cells were retained for downstream analysis. Other considerations are whether single cells have actually been isolated or whether indeed two or more cells have been mistakenly assessed in a particular sample. The scale and capabilities of single-cell RNA-sequencing methods have expanded rapidly in recent years, enabling major discoveries and large-scale cell mapping efforts. Baran, Y. et al. Nevertheless, the use of UMIs can significantly reduce amplification bias and therefore improve precision [32]. Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. For example, resected tumours might be routinely assessed for the presence of rare malignant and chemo-resistant cancer cells. Although scRNA-seq is now more accessible to first-time researchers through commercial reagents and platforms, this is less true for the crucial bio-informatic and computational demands of a scRNA-seq study. Cell 174, 716729 (2018). A comparison of automatic cell identification methods for single-cell RNA sequencing data. It is important to note that commercial kits and reagents now exist for all the wet-lab steps of a scRNA-seq protocol, from lysing cells through to preparing samples for sequencing. Typically, the most biologically interesting heterogeneity among cells, other than heterogeneity in lineage identity, is due to different intermediate transcriptional states, which can provide information about whether the regulation of individual cells is normal or aberrant. Here, we consider what the next 5years might hold for scRNA-seq from the perspective of clinical and experimental researchers looking to use this technology for the first time. Bacher, R. et al. Moreover, downscaling the reactions to nanoliter volumes has been shown to improve detection sensitivity [33] and quantitative accuracy [44]. The .gov means its official. FEBS Lett. 2015;33:15560. Stein, C. K. et al. We are grateful to Valentine Svensson for useful discussions during the preparation of this manuscript. The dropout events . Clustering analysis is critical to transcriptome research as it allows for further identification and discovery of new cell types. We also thank S. Ballerau, M. Bttner, M. Do Nascimentoo Lopes Primo, J. Lee, R. Lyu, E. Madissoon, R. Martinez Nunez, S. Y. Mller, K. Polanski, P. Qiao and J. Westoby for their contributions to teaching the course and developing the material. Provided by the Springer Nature SharedIt content-sharing initiative. Hafemeister, C. & Satija, R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. Transcriptomics was initially conducted on ensembles of millions of cells, firstly with hybridization-based microarrays, and later with next-generation sequencing (NGS) techniques referred to as RNA-seq. Single-cell RNA sequencing (scRNA-seq) technologies allow the dissection of gene expression at single-cell resolution, which greatly revolutionizes transcriptomic studies. McGinnis, C. S., Murrow, L. M. & Gartner, Z. J. DoubletFinder: doublet detection in single-cell RNA sequencing data using artificial nearest neighbors. government site. Nature, 501, 338-345. doi: 10.1038/nature12625. Science. However, with the increasing commercial availability of scRNA-seq platforms, and the rapid ongoing maturation of bioinformatics approaches, a point has been reached where any biomedical researcher or clinician can use scRNA-seq to make exciting discoveries. This includes the study of monoallelic gene expression [9, 26, 27], splicing patterns [12], as well as noise during transcriptional responses [7, 12, 13, 28, 29]. Nat Commun. In bulk RNAseq, we measure the average expression of . t-SNE t-distributed stochastic neighbour embedding. Genome Medicine In this article and our companion website (https://scrnaseq-course.cog.sanger.ac.uk/website/index.html), we provide guidelines regarding best practices for performing computational analyses. Diffusion maps for high-dimensional single-cell analysis of differentiation data. Haque A, Engel J, Teichmann SA, Lnnberg T. Genome Med. Trends Genet. et al. A scaling normalization method for differential expression analysis of RNA-seq data. An official website of the United States government. Because of this imperfect coverage, the commonly used unit of normalized transcript levels used for bulk RNA-seq, expressed as reads per kilobase per million (RPKM), is biased on a single-cell level, and instead the related unit transcripts per million (TPM) should be used for scRNA-seq [66]. Butler, A., Hoffman, P., Smibert, P., Papalexi, E. & Satija, R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Given that most scRNA-seq data are generated by sequencing cDNA libraries from single cells that are barcoded and pooled, the depth of single-cell sequencing (that is, the number of transcripts detected from each cell) diminishes as the number of libraries included in a sequencing run is increased, owing to a finite sequencing capacity per run. Droplet barcoding for single-cell transcriptomics applied to embryonic stem cells. 2011;9:724. Nat Methods. Front Genet. Unique molecular identifier A variation of barcoding, in which the RNA molecules to be amplified are tagged with random n-mer oligonucleotides. Massively parallel digital transcriptional profiling of single cells. PubMed 2015;163:141327. & Yosef, N. Deep generative modeling for single-cell transcriptomics. Nat Biotechnol. BMC Genomics 19, 477 (2018). While scRNA-seq workflows are conceptually closely related to population-level transcriptomics protocols, data from scRNA-seq experiments have several features that require specific bioinformatics approaches. In summary, an understanding of the bioinformatic and computational issues involved in scRNA-seq studies is needed, and specialist support for biomedical researchers and clinicians from bio-informaticians who are comfortable with handling scRNA-seq datasets would be beneficial. Cell-size variation can also be closely related to proliferative status and cell-cycle phase. Zheng, G. X. Y. et al. Nat. Nat. 2014;509:3715. PubMed Central Stubbington MJ, Lnnberg T, Proserpio V, Clare S, Speak AO, Dougan G, et al. Zhang, X. et al. 24, 496510 (2014). Lun, A. T. L., Calero-Nieto, F. J., Haim-Vilmovsky, L., Gttgens, B. With these increasing opportunities for single-cell transcriptome characterization, we have witnessed remarkable diversification of experimental protocols, each coming with characteristic strengths and weaknesses. Bookshelf Ziegenhain, C. et al. Sasagawa Y, Nikaido I, Hayashi T, Danno H, Uno KD, Imai T, et al. Brief Bioinformatics 20, 15831589 (2019). sensitive highly-multiplexed single-cell RNA-Seq. The crux of the issue is how to examine tens of thousands of genes possibly being expressed in one cell, and provide a meaningful comparison to another cell expressing the same large number of genes, but in a very different manner. Cell Rep. 2012;2:66673. official website and that any information you provide is encrypted A comparison of single-cell trajectory inference methods. J Clin Med. Single-Cell RNA-Seq Library Preparation and Sequencing. Next, as an extension to a full blood count, scRNA-seq assessments will provide in-depth information on the response of immune cells, which again will inform diagnoses and the choice of therapy. Brief Funct. BioRxiv. Data denoising with transfer learning in single-cell transcriptomics. Ronan T, Qi Z, Naegle KM. Cell Syst. Freytag, S., Tian, L., Lnnstedt, I., Ng, M. & Bahlo, M. Comparison of clustering tools in R for medium-sized 10 Genomics single-cell RNA-sequencing data. Nat Genet. Genome Biol. 2008, P10008 (2008). Highly parallel genome-wide expression profiling of individual cells using nanoliter droplets. Analysis of allelic expression patterns in clonal somatic cells by single-cell RNA-seq. These trajectory-inference methods are conceptually based on identification of intermediate cell states, and the most recent tools are able to trace both linear differentiation processes as well as multipronged fate decisions [22,91,92,93,94,, 24, 9095]. Macaulay IC, Ponting CP, Voet T. Single-cell multiomics: multiple measurements from single cells. We discuss some of the most common tasks and the tools available for addressing central biological questions. Kar G, Kim JK, Kolodziejczyk AA, Natarajan KN, Torlai Triglia E, Mifsud B, et al. Methods 85, 5461 (2015). In BioRxiv. By using this website, you agree to our Genome Biol. urauskien J, Yau C. pcaReduce: hierarchical clustering of single cell transcriptional profiles. Kobak, D. & Linderman, G. C. UMAP does not preserve global structure any better than t-SNE when using the same initialization. Traditional next-generation sequencing (NGS) examines the genome of a cell population, such as a cell culture, a tissue, an organ or an entire organism.Its output is the "average genome" of the cell population.