Single cell RNA sequencing: modern approaches and achievements

Full Text

Introduction

The human body is a complex system consisting of more than 400 different cell types and more than 3×10¹³ unique cells in its composition [1]. Despite almost all cells in the human body sharing the same set of genes, their transcriptome function in each type and each individual cell exhibits a unique activity for only a certain part of these genes, which increases the relevance of the issue of studying their unique properties and heterogeneity.

Sequencing technology became one of the common tools for studying cell heterogeneity, including genomics, transcriptomics, epigenomics, proteomics and metabolomics. Sequencing allows to establish the primary structure of the studied linear biopolymers, for DNA and RNA it allows to establish their nucleotide sequence, presented in simple text form. During its widespread use over the past 50 years, sequencing technology has become mundane in various fields of biology and medicine, and currently represents three main generations of this technology [2, 3] (presented in Table 1, [4–15]). The most frequently employed set of methods belongs to the second generation sequencing, which is also called short-read sequencing or next generation sequencing (NGS) is currently the “gold standard” in scientific and laboratory genetic research [3].

Table 1
Generations of sequencing methods

Generation and characteristics	Examples of technologies	Ref.
First generationSanger sequencing, used from 1975 to the present.Average read length: 600–1000 base pairs, analysis of single samples	Manual and automated protocols using radioactively or fluorescently labeled reagents and gel electrophoresis	4, 5
Second generationNext generation sequencing (NGS) or short-read sequencing or High-throughput sequencing, used from 2005 to the present.Short read length: 250–800 base pairs, multiple parallel analysis.	“Ion Torrent”, Thermo Fisher Scientific, USA	6, 7
	“Illumina”, Illumina, Inc., USA	8, 9
	“MGI/BGI”, Beijing Genomics Institute, BGI, China	10, 11
Third generationLong-read sequencing, under development and implementation since 2010.High read length: 105–107 base pairs, single and multiple parallel analysis.	“PacBio”, Pacific Biosciences, USA	12, 13
	“ONT”, Oxford Nanopore Technologies, UK	14, 15

The main limitation of the basic sequencing approach is its non-specific character. In its classical implementation, the second-generation sequencing technology provides only average data (bulk sequencing) for the total populations of the analyzed cells; data for individual cell populations, represented in insignificant quantities, are then “lost” against the background of the total mass of recorded data. This limitation gave an impetus to the development of a technology for isolating and indexing unique cells for their subsequent mass analysis using NGS methods. Over the past fifteen years, classical approaches to RNA sequencing (bulk RNA-seq) based on cell populations had a significant role in deciphering transcriptome variations across the genome in a wide range of areas, including oncology, cell development and aging, and cellular homeostasis [16–18]. However, since bulk RNA-seq data, as described above, represent only an “average” of gene expression in individual cells, they may obscure transcriptional trends of individual subpopulations with the most common cell types or states [19].

A modification of NGS technology in the form of single-cell RNA sequencing (scRNA-seq) overcame this obstacle, providing a broad range of possibilities for gene expression profiling at single-cell resolution. With scRNA-seq, it is now possible to analyze the transcriptome at the single-cell level for more than millions of cells in a single study. Most importantly, the development of the method has significantly reduced the cost while improving automation and throughput. This has enabled various discoveries in various fields, such as identifying new cell types [20, 21], understanding a range of cellular dynamic processes [22–25], revealing intracellular gene variations [26], and RNA processing [27]. As a result, scRNA-seq has begun to be used as a key tool in the field of precision medicine [28, 29].

Concept of scRNA-seq method

Since its first application in 2009 [30], scRNA-seq technology has opened the possibility of studying the underlying cellular heterogeneity of complex systems [31]. However, at the time of its introduction, practical procedures were labor-intensive and expensive, presented as “closed solutions” in the form of reagent kits and protocols from a single manufacturer, requiring the researcher to have a full product line of equipment from a specific manufacturer. Currently, the application of the technology is relatively unified and can be successfully applied in typical laboratories, which promotes scRNA-seq to be used as a standard procedure in an increasing number of tasks.

The main concept of scRNA-seq technology has not changed since its discovery and represents approaches for isolating individual cells, capturing and scaling their genetic material with subsequent use of NGS sequencing [32, 33]. Later, various modifications of scRNA-seq technology were proposed, supported by the development and evolution of methods for bioinformatics analysis of the final data. All scRNA-seq technologies can be described in the form of five key steps presented in Figure 1: isolation of single cells from biomaterial and their combination with reagents, isolation of the transcriptome from a single cell and its capture with a reagent, individual barcoding of the transcriptome by reverse transcription (synthesis and amplification of cDNA), library generation and NGS sequencing, additional bioinformatics analysis of the data [34]. The first three steps of the technology are the most difficult to implement, the remaining two are true for all sequencing methods.

Single cell isolation is a mechanical process of capturing cells after their dissociation from tissues and combining them into separate isolated cavities along with reagents. With classical sequencing, the transcriptome can only be examined at a general level and cannot distinguish individual cell variations. In the case of scRNA-seq technology, special attention is required to obtain the genetic material from each cell in a unique, isolated form and preventing mixing.

Fig. 1. Schematic of the scRNA-seq experiment. A mixture of beads, cells, and reagents is pumped through an orthogonal fluidic system to form droplets via a reverse emulsion (a). The system parameters are adjusted to maximize the ratio of single cell captures per droplet while minimizing double captures (b). Each droplet is an individual mini-reaction chamber containing all the necessary reagents for cell lysis, reverse transcription, and amplification (c), thus generating large quantities of barcoded DNA complementary to the cellular transcriptome, suitable for subsequent NGS sequencing (d). Each individual bead contains on its surface numerous complex primers, including in their composition both identical regions necessary for, for example, transcription or amplification (“T/A Primer”), unique for each specific bead regions for cell barcoding (“Barcode”) and unique for each primer regions for barcoding each specific transcriptome sample extracted from the cell (“UMI”). The terminal region of the primer contains a sequence of oligonucleotides for complementary capture of the type of transcriptome of interest, for example, poly-dT regions for binding to many RNA variants. After such capture, conditions are created for reverse transcription and obtaining cDNA (e) and further, cycles of synthesis of the complete barcoded DNA variant (f).
Note: PCR — polymerase chain reaction; IVT — in vitro transcription; UMI — unique molecular identifier; cDNA — complementary DNA.

Methods for isolating and capturing single cells vary greatly depending on the organism, tissue, or cell properties [35]. Cell isolation can be achieved for whole cells or, for example, individual elements of cells [36]. The most common methods for isolating and capturing single cells include microfluidic systems, Fluorescence-Activated Cell Sorting (FACS) or Magnetic Activated Cell Sorting (MACS), and laser microdissection [31]. The main goal of each method is to perform a single capture into an isolated reaction mixture, which will also contain reagents for the further process.

The process can be schematically considered using the example of a microfluidic system shown in Figure 1. A suspension of dissociated cells is fed to a microfluidic system with selected parameters of channel sizes and flow rates; a reaction mixture in an aqueous medium and oils are fed to orthogonal directions of the chip to form spherical capsules — droplets according to the principle of inverse emulsion (Figure 1a). With the correct selection of system parameters, conditions are achieved where the majority of single cells are enclosed in emulsion spheres, which contain only one cell and a solution of reagents for further processing, a significant part of the empty spheres and only a small part of the spheres with double inclusion of cells (Figure 1b). Each of the spheres will represent a separate reaction vessel, ensuring the isolated flow of processes. In this way, unique genetic and biochemical information is extracted for its further transformation and study [37].

Transcriptome isolation and capture occurs due to the presence of a lysis buffer and special microparticles in the reaction mixture. The lysis buffer ensures cell lysis and the release of genetic material into the reaction mixture, the microparticle captures genetic material from the reaction mixture via oligonucleotide molecules grafted onto the surface — primers (Figure 1c).

The general structure of the primer can be divided into at least three main sections: (1) cell barcode (Cell Label / Barcode) — a short sequence common to all primers on one microparticle, for the precise identification of all transcripts belonging to the same cell; (2) unique molecular identifier (Unique Molecular Identifier, UMI) — a molecular tag for the precise identification of each read of each unique captured transcript and their number before the amplification stage [38]; (3) a capture and amplification segment designed for a specific capture target, e. g. the basic version of the capture segment for RNA is a poly-T tail to capture the 3’ end of the transcript (Figure 1e). In modern scRNA-seq methods, additional sections of the primer structure may be present, e. g. for cDNA amplification or further work in a specific NGS sequencing system. Thus, a specific part of the genetic material released into the reaction mixture is captured by a microparticle designed for this type of genetic material due to complementary interaction.

Barcoding of the single-cell transcriptome is a key step of the scRNA-seq technology in all available protocols and is the main difference from mass RNA-seq. The introduction of this step allows for the unambiguous designation of each transcriptome with a unique cellular barcode. The transcript complementarily bound to the microparticle is subjected to several cycles of amplification using the polymerase chain reaction (PCR) or in vitro transcription (IVT) due to the presence of the necessary reagents and primers in the reaction mixture. In the first amplification cycle, DNA complementary to the transcript on the microparticle is synthesized, forming a single chain with a complex primer, and, as a result, the barcode described above (Figure 1f). Then, in subsequent amplification cycles, multiple cDNA synthesis occurs in the reaction mixture [37].

The subsequent steps of the technology are similar to classical NGS sequencing. Library preparation and quality control, the sequencing process itself and computer processing of the obtained data are carried out in accordance with the protocol and using reagents from the sequencing equipment manufacturer (Figure 1d). Historically, the most popular sequencing platforms in the majority of cases are various series of NGS sequencers from Illumina (Illumina, Inc., USA) due to the high efficiency and high quality of results obtained using this equipment. Most of the commercially available scRNA-seq solutions presented have been developed and adapted for Illumina equipment. A relatively new player in this field is the MGI/BGI (Beijing Genomics Institute, BGI, China) line of NGS sequencers, which is competitively increasing its market share and demonstrating equivalent sequencing results even in studies of individual cells [39]. Besides, the growth and development of software should not be overlooked. Recent advances enable additional processing and visualization of data obtained by the scRNA-seq method.

The process of data processing in the scRNA-seq experiment, unlike bulk analysis options, requires the use of additional tools to solve emerging problems due to the increase in the amount of individual data [40, 41]. Currently, more than 600 stand-alone tools in the R and Python programming languages are available for analyzing and exploring scRNA-seq data [42], but most of them have a high entry threshold and will be difficult for beginners to master [43, 44]. Furthermore, due to the relative novelty of the method, working with different types of cells and biological samples may require a different approach to data processing and analysis, which also complicates working with scRNA-seq data.

The obtained sequencing data undergo primary processing by the algorithms of the software of the sequencing equipment manufacturer. Then the data undergoes additional processing using specialized scRNA-seq software packages. The data are presented in the form of gene expression matrices (in the “cell — gene” format), where the number of copies of the detected transcripts is indicated for each gene. Such separation of cells is achieved by introducing additional identifiers at the barcoding stage — Cell Label / Barcode and UMI. The presence of the Cell Label / Barcode allows to associate all readings with a specific cell, and the presence of unique molecular identifiers UMI allows to count the number of readings of specific transcripts for each gene within a single cell. Further processing involves dividing cells into populations using normalization methods (SCT / log), and then reducing the dimensionality using the Principal Component Analysis (PCA) method and the nonlinear dimensionality reduction (Uniform Manifold Approximation and Projection, UMAP) method. This allows to plot single cells on a two-dimensional graph considering their differences in gene expression for their most variable representatives. After this, individual cell clusters are isolated by detecting cells with a similar expression profile. The isolated clusters are then identified as individual cell populations with a specific phenotype (Figure 2).

Fig. 2. Data processing in an scRNA-seq experiment. Various variants of heterogeneous cell populations can be studied: a mixture of different cell types, such as blood; a tissue with a gradient of different cells that may be associated during development; cells transitioning from one state to another, such as differentiating cells (a). The main sources of variation in the experiment are technical and biological. Technical noise is estimated and corrected for using approaches that relate the number of unique reads and labels. Sources of biological variability can be variability among different subpopulations, cell-to-cell variability in gene expression and transcription kinetics, and variability due to cell function and biological processes, such as the cell cycle, which serves as a source of data for further analysis (b). Identification of cell populations can be performed using principal component analysis (PCA) or hierarchical clustering. There are also various approaches to characterize subpopulations: searching for cell type markers by analyzing differential expression between different groups of cells; frequency of occurrence of cell populations; identification of genes that exhibit certain patterns during a process, such as development or response to stimuli: genes which expression either increases or decreases throughout the process, but, most interestingly, genes that are expressed transiently in intermediate cell types, since these genes may be important for the process (c).
Note: UMI — unique molecular identifier.

A classic example of a software package for working with scRNA-seq data regardless of the cell types studied is the approach described in [45], which proposes four ready-to-use software packages that include raw count normalization, feature selection, dimensionality reduction, and clustering, shown in Figure 2. The proposed software packages use both R and Python programming languages and include Seurat (R) [46], Scanpy (Python) [47], Monocle (R) [48, 49], and gf-icf (R) [50]. Using this approach will allow inexperienced users to analyze scRNA-seq data without complications and loss of information. An example of using this approach with a description of the steps, demonstration of graphs and command lines is available on “github” [51].

Examples of using the scRNA-seq method

During their development and application over the past 15 years, scRNA-seq technologies have grown into various solutions, including commercial ones (presented in Table 2, [52–68]), which can be classified in various ways, the most relevant of which, in our opinion, is by the cell isolation method. Similar to what was described in the section above, the technologies can be divided into microfluidic options that use chips and the formation of emulsion microdroplets for separation, options with magnetic or fluorescent sorting, and other options that use, for example, laser dissection or cytometry. For example, in the field of scRNA-seq, microfluidic technologies have recently gained popularity due to their cost-effectiveness, high performance, and moderate requirements for the budget and the size of the generated data [27, 28]. Another important parameter is the division into technologies with high throughput but low detection limits, and, conversely, low throughput but deep detection limits.

Table 2
scRNA-seq technologies

Technology	Isolation type	Add. parameters	Year / Ref.	Examples of commercial technologies
Smart-seq	FACS	PCR, Full-length	2012 [52]	SMART-Seq®, Takara Bio, Japan [62]
Smart-seq2	FACS	PCR, Full-length	2013 [53]
MATQ-seq	FACS	PCR, Full-length	2017 [54]
CEL-seq	FACS	IVT, 3’ end	2012 [55]
MARS-seq	FACS	IVT, 3’ end	2014 [56]
Fluidigm C1	Micro-fluidic	PCR, Full-length	2013 [57]	BD Rhapsody®, BD Biosciences, USA [63]Quantum Scale ®, Scale Bioscience, USA, [64]Gexscope®, Singlerion, Germany [65]
Seq-Well	Micro-fluidic	PCR, 3’ end	2017 [58]
Drop-seq	Microdroplets	PCR, 3’ end	2015 [37]	Chromium®, Gem-X®, 10X Genomics, USA [66]DNBelab® C Series, MGI/BGI, China [67]ddSEQ® Single-Cell, BioRad, USA [68]
inDrop-seq	Microdroplets	IVT, 3’ end	2015 [59]
10x Genomics	Microdroplets	PCR, 3’ end	2016 [60]
DNBelab C4	Microdroplets	PCR, 3’ end	2019 [61]

Note: FACS — fluorescence-activated cell sorting; PCR — polymerase chain reaction; IVT — in vitro transcription.

The most well-known commercial technologies currently in use are Chromium® from 10X Genomics, USA and BD Rhapsody® from BD Biosciences, USA, which provide users with a wide range of commercial solutions and equipment for sample preparation using scRNA-seq technology and compatible with NSG sequencing methods from various manufacturers. Technologies have successfully competed for a long time and are subject to constant comparison [69]. As scRNA-seq technology is mastered and distributed, more and more manufacturers are mastering the technology and offering their commercial solutions, for example, ddSEQ® Single-Cell from BioRad, DNBelab® C Series from MGI/BGI, China, Quantum Scale® from Scale Bioscience, USA and Gexscope® from Singlerion, Germany.

However, the implementation of scRNA-seq methods is still not widely available for many laboratories due to its relatively high cost [34, 35]. At the moment, the technology is still relatively expensive, the cost of a reagent kit varies from several thousand to several hundred US dollars excluding the cost of equipment that performs operations according to the protocol. Nevertheless, it is the presence and development of competition between commercial scRNA-seq solutions that has a positive effect on this gradual reduction in the cost of using the technology. Moreover, a number of companies have already announced the development and imminent release of new cost-effective commercial solutions for scRNA-seq technology. Organizations such as Parse Biosciences, USA and Fluent BioSciences, USA announce the development of solutions that allow to scale up analysis multiple times and simplify the procedure for analyzing and processing data due to the implementation of new technologies and use of machine learning and artificial intelligence. ArgenTAG, Argentina, in turn, is preparing a commercial solution for applying scRNA-seq to third-generation sequencers, in particular Oxford Nanopore, and plans to reduce the cost due to both more productive cell barcoding technologies and the advantage of long reads of new sequencers. Also, Scipio Bioscience, France, has proposed an alternative approach to barcoding in hydrogel without the use of expensive equipment and is preparing its kits for release to the market. Such a solution will allow scRNA-seq experiments to be carried out simply at the lab table further reducing their cost.

Below are examples of the application of the technology and the results obtained in the fields of clinical diagnostics, cellular aging and oncology, obtained mainly using only commercial solutions Chromium® from 10X Genomics, USA and BD Rhapsody® from BD Biosciences, USA. The rapid development of scRNA-seq technology inspires optimism and promises growth in the prospects for future medical and biological research.

Oncology

Currently, scRNA-seq technology is widely used in oncology for profiling the transcriptomes of individual cells. The data obtained using this method have improved our understanding of the biological characteristics and dynamics within cancer lesions, including information related to malignant cell and immune cell landscapes, tumor heterogeneity, circulating tumor cells, and the underlying mechanisms of tumor biological behavior [70]. Various scRNA-seq technologies have been used to study cell hierarchy and decipher clonal heterogeneity in normal bone marrow and acute myeloid leukemia [71], and to search for new diagnostic and prognostic biomarkers among various tumors [72], such as bladder cancer and pancreatic cancer [73]. Systematic accumulation of scRNA-seq data has allowed the creation of a cancer cell atlas that covers a spectrum of cancer types [74–84] and may provide guidance for treatment of conditions where basic guidelines are lacking [85].

The use of scRNA-seq has enabled the characterization of dynamic interactions within a single cell by analyzing transcriptome variation over space and time. The sequential attachment of spatial barcodes allows the encoding and extraction of location information of individual cells, which can effectively inform research and diagnostics. Thus, tracking temporal and spatial variability allows the study of the impact on tumor heterogeneity and stress response, which in turn can be used for cancer diagnosis, classification, and treatment [86–88]. For example, a study [89] deciphered a detailed spatial map of individual cell phenotypes and cell systems, which the researchers used to demonstrate the phenotypic heterogeneity of tumor cells in breast cancer. Similar results were achieved in studies [90, 91] on pancreatic and breast cancer.

The use of scRNA-seq has enabled the identification of gene expression changes during cancer progression. Studies using single cell sequencing of adjacent normal tissues and adenomas at different stages in patients have comprehensively revealed genomic changes, clonal architecture, and metabolic dynamics during tumor development, which in turn has provided insight into tumor progression inhibition [92]. For example, a study [93] investigated the cell evolutionary pathway in salivary gland squamous cell carcinoma, including the stages of basal cell carcinogenesis, activation of Wnt signaling cascades, and subsequent cell differentiation into luminal cells. Another example is a study [94] describing the mechanism of transformation of familial adenomatous polyposis to adenocarcinoma, indicating that malignant cells retain epithelial characteristics while undergoing rapid migration in breast cancer.

The use of scRNA-seq allows clustering of different cells. Clustering of tumor cell subsets helps to identify rare subsets or a spectrum of cell states of malignant cells [70]. For example, studies [95–99] using scRNA-seq have identified melanoma cells in intermediate states and malignant cell subsets with unique characteristics in pancreatic, ovarian, and gastric cancers. scRNA-seq can also be used to cluster subsets of immune cells that make up the tumor immune microenvironment, also allowing to assess the impact on immunotherapeutic response and clinical outcomes [70]. For example, a study [100] describes gradual replacement of resident myeloid cells by macrophages and monocyte-derived DCs upon T cell depletion in metastatic lung adenocarcinoma.

Senescence

The field of studying cellular senescence is an interesting example of the early use of scRNA-seq technology. Various factors can induce cellular senescence, and the rate of accumulation of senescent cells varies across organs. In addition, since heterogeneity is due to the spatiotemporal context of senescent cells, in vivo studies are required to improve the understanding of senescent cells [101]. While previous attempts have been made to find markers commonly expressed in senescent cells, researchers have been actively using bulk RNA-seq as a primary tool to study cells and their relationships over time [102–105].

The use of scRNA-seq in senescence is currently a relatively new approach to study rare senescent cells. This is because only modern scRNA-seq technologies achieve the minimum required throughput of ~10,000 cells per approach to work with rather rare senescent cells in the bulk, which constitute only 2% of specific tissues in vivo [106]. Technologies such as InDrops and Drop-seq can easily scale up cell numbers, thereby increasing the ability to detect rare cells [107]. Therefore, improving the throughput of scRNA-seq and ways to enrich for rare senescent cells are priority areas for development.

One of the most striking examples of the use of scRNA-seq in this area is the creation of the Tabula Muris Senis — atlas of mouse aging [108], where various tissues of male and female mice were analyzed over time. The study observed the emergence of aging-specific cell types in tissues such as the liver and bladder, as well as immune infiltration of various tissues. It was noted that the number of T and B cells in gonadal adipose tissue increased with age, and a cluster of B cells highly expressing immunoglobulin J was found specifically in old mice. Moreover, the use of the scRNA-seq data sets by the researchers [108, 109] showed a certain discrepancy between the expected results and the already established theory about the expression of the aging marker p16. In the study [108], it was noted that the fraction of cells expressing p16 doubled after 24 months compared to 3 months, and the level of p16 expression also increased approximately 2‑fold. However, according to the study [109], the proportion of p16‑expressing cells and the average level of p16 expression did not increase during aging. This allowed to conclude both that p16‑expressing senescent cells are quite rare and are likely to be lost during tissue dissociation and sequencing library creation, and that the pattern of accumulation of p16‑expressing senescent cells may differ between subjects [101].

For ethical reasons, scRNA-seq analysis for human tissues has been performed only in limited quantities. Several studies [110–115] have obtained data for the creation of aging cell atlases for muscle, pancreas, brain, retina, and blood cells. A study [116] using scRNA-seq on human lung tissue was able to show a relationship between senescent cell populations and age-related progression of pulmonary fibrosis, despite the poor representation of classical senescent markers determined by bulk RNA-seq. These data led to the conclusion that in vivo human senescent cells may have a different transcriptomic signature from senescent cells in vitro and the use of single-cell spatial transcriptomics to study different tissues longitudinally may help to expand the understanding of the systemic effects of senescent cells [117].

Diagnostics and clinical research

The use of scRNA-seq technology paired with machine learning is a powerful tool for creating complex diagnostic systems for various diseases. For example, in the study of systemic lupus erythematosus [118], a machine learning-based pipeline was presented that was able to compare hundreds of single-cell proteome and transcriptome data points and identify several working combinations of biomarkers for diagnosis and further monitoring of the disease.

A study [119] devoted to postmenopausal osteoporosis also undertook solving the problem of discovering and screening new biomarkers for the disease, as well as attempts to identify new therapeutic targets by combining scRNA-seq and bulk RNA-seq technologies. Three potential biomarkers were identified and a correlation with the effectiveness of several potential drugs for treatment was established. These data were used for early diagnosis and targeted treatment of osteoporosis.

Another study [120] using a combination of scRNA-seq and bulk RNA-seq technologies aimed at finding biomarkers and developing methods for diagnosing sepsis. A specific biomarker has been identified that demonstrates diagnostic and prognostic value in sepsis and has been nominated for potential use in the treatment of this pathology. Single-cell sequencing is widely used to measure the clinical efficacy and safety of new drugs in clinical trials. Single-cell sequencing can be used to study dynamic regulatory networks and identify heterogeneous cellular behavior in response to different chemotherapy or radiation in cell lines or tumor samples such as esophageal squamous cell carcinoma [121]. As an example, a study [122] described the sci-Plex method used to decipher transcriptomic changes and their underlying mechanisms in millions of malignant cells in response to drug therapy.

Conclusion

The use of the scRNA-seq method allows to obtain sequencing data with greater accuracy, volume and quality. This opens up a wide field to conduct a detailed study of biological processes occurring in the central and peripheral organs of the immune system, inflammatory and tumor processes, as well as cellular aging processes, which is already confirmed by examples of the conducted studies.

In our opinion, mastering this method is one of the current professional challenges for domestic research teams. Tackling such a challenge will require increasing the level of technical equipment, gaining experience in using the current and previous versions of the technology — bulk RNA-seq / NGS, and studying modern bioinformatics tools for processing and interpreting data, which in turn implies the development of competencies in the field of immunology and biology, as well as the study and implementation of programming languages in practice.

Based on the scientific examples presented in this review, it can be concluded that the development and successful application of the scRNA-seq method in scientific and clinical practice will become the key to a wide range of future discoveries and successful accurate personalized diagnostics and healthcare.

About the authors

Artem E. Gusev

National Research Center Institute of Immunology of the Federal Medical-Biological Agency

Email: kofiadi@mail.ru
ORCID iD: 0009-0002-5810-098X
SPIN-code: 4385-0589
Moscow, Russian Federation

Petr V. Chernov

National Research Center Institute of Immunology of the Federal Medical-Biological Agency

Email: kofiadi@mail.ru
ORCID iD: 0009-0005-0642-8723
Moscow, Russian Federation

Nikolai A. Dmitriev

National Research Center Institute of Immunology of the Federal Medical-Biological Agency

Email: kofiadi@mail.ru
ORCID iD: 0009-0002-8381-8512
Moscow, Russian Federation

Ilya A. Kofiadi

National Research Center Institute of Immunology of the Federal Medical-Biological Agency

Author for correspondence.
Email: kofiadi@mail.ru
ORCID iD: 0000-0001-9280-8282
SPIN-code: 5730-0925
Moscow, Russian Federation

References

Supplementary files