Single-Cell Genomics: From Concept to Application

Single-cell genomics is the study of genomes, transcriptomes, epigenomes, and other molecular features at the resolution of individual cells instead of bulk tissue samples. This field emerged from the recognition that population-averaged measurements obscure the heterogeneity present in every biological sample. For researchers, analysts, and life-science professionals new to this area, single-cell genomics provides tools to resolve cell types, identify rare populations, trace developmental trajectories, and link molecular states to clinical or agricultural outcomes. This article explains the core technologies, the analytical workflow, practical implementation considerations, and the limitations that shape how single-cell data should be interpreted and reported.

What Single-Cell Genomics Measures

Single-cell genomics encompasses multiple molecular modalities. The most established approach is single-cell RNA sequencing (scRNA-seq), which quantifies gene expression in individual cells. Since the first single-cell RNA-sequencing study appeared in 2009, the field has expanded to include single-cell genome sequencing for detecting somatic mutations, single-cell epigenomics for chromatin accessibility and methylation patterns, single-cell proteomics, and single-cell metabolomics [6][9]. These technologies have moved horizontally across molecular layers and vertically toward integrating multiple omics data types within the same cell [9].

The core rationale for single-cell resolution is that bulk measurements average signals across millions of cells, masking heterogeneity and rare populations [26]. A tumor sample analyzed in bulk might show an intermediate expression value for a marker gene, obscuring the fact that only a small subpopulation of cells expresses that marker at high levels. Single-cell methods preserve this information by assigning molecular measurements to individual cells.

Single-cell genomics has had major impacts in microbiology, neurobiology, cancer research, and developmental biology [6]. In cancer, single-cell approaches reveal intratumoral heterogeneity, clonal evolution, and the composition of the tumor microenvironment [18]. In neuroscience, single-cell and spatial technologies have enabled the construction of brain atlases across species, identifying genes, cell types, and molecular pathways involved in brain development and function [13]. In immunology, single-cell analysis characterizes immune cell states and cell-cell communication networks [8]. In aging research, single-cell sequencing has identified aging biomarkers, uncovered novel aging cell subpopulations, and revealed tissue-specific aging characteristics [11].

The technology has also reached veterinary and agricultural science. Single-cell genomics methods have been applied to livestock species, including the construction of a skeletal muscle atlas in pigs that identified cell-type-specific genetic regulators of meat production traits [16]. A dedicated review of single-cell genomics in veterinary science covers methods and applications relevant to animal health and production [27].

Core Technologies and Their Tradeoffs

Single-Cell RNA Sequencing

Single-cell RNA sequencing measures the transcriptome of individual cells. The workflow begins with cell isolation, followed by reverse transcription, amplification, and library preparation, then sequencing and computational analysis [5]. Commercial platforms have standardized much of this process, and improvements in reliability have opened the technology to clinical laboratories [6].

The choice of platform affects throughput, cost, sensitivity, and the types of questions that can be answered. Some platforms capture thousands to millions of cells per experiment, while others are designed for smaller numbers of cells with deeper coverage. The rapid evolution of these technologies means that researchers must evaluate current options against their specific biological question [12].

Single-Cell Genome Sequencing

Single-cell genome sequencing detects DNA alterations, including somatic mutations and copy number variations, in individual cells. This approach can reveal clonal dynamics and the evolutionary relationships between cells within a sample [10]. Applications include cancer genomics, where understanding intratumoral genetic heterogeneity informs treatment strategies, and forensic science, where single-cell DNA profiling can resolve mixed stains that bulk methods cannot separate [21].

Single-Cell Epigenomics

Single-cell epigenomic methods measure chromatin accessibility, DNA methylation, and histone modifications at single-cell resolution. These approaches reveal regulatory mechanisms that shape cell identity and response to stimuli. Integration of epigenomic data with transcriptomic data from the same cells provides a more complete picture of gene regulation [9][11].

Spatial Transcriptomics

Standard single-cell methods require tissue dissociation, which destroys spatial context. Spatial transcriptomics technologies, including MERFISH, CosMx, Stereo-seq, and Visium, profile gene expression within intact tissue architecture [13]. These methods map gene expression to anatomic locations, enabling the characterization of cellular heterogeneity across tissue regions and the spatial organization of cellular circuits [13]. Integration of spatial data with single-cell sequencing data has become a standard approach in many studies [8][13].

Multi-Omics Integration

Recent technologies measure multiple molecular layers from the same cell. These multimodal approaches combine transcriptomics with epigenomics, proteomics, or other data types [9][12]. Multi-omics integration provides a more complete view of cellular state and enables researchers to link genetic variation to gene expression to protein abundance within individual cells.

The Single-Cell Genomics Workflow

Experimental Design

The success of a single-cell genomics project depends on decisions made before any cells are processed. Key considerations include the biological question, the choice of modality, the number of cells needed, the depth of sequencing, and the number of biological replicates [10][12]. Single-cell experiments are expensive, and inadequate replication can undermine statistical power. Researchers should consider the expected frequency of rare cell types, the need for trajectory inference, and the requirements of downstream analyses when determining cell numbers.

Sample preparation is a critical source of variability. Most single-cell RNA sequencing protocols require fresh, high-quality single-cell suspensions processed immediately to preserve transcriptional profiles [15]. This constraint complicates studies involving remote collection sites or long preparation times. Commercial preservation methods, including fixation and cryopreservation, now allow samples to be collected at the point of care and processed months later [15]. A multisite study by the Association of Biomolecular Research Facilities evaluated the performance and reproducibility of three such platforms, providing guidance on their use [15].

Cell Isolation and Library Preparation

Cell isolation methods include fluorescence-activated cell sorting, microfluidics, droplet-based platforms, and nanowell technologies. The choice of method affects cell capture efficiency, doublet rates, and the ability to process specific sample types [5][12]. For tissues that are difficult to dissociate, such as adipose tissue or brain, single-nucleus RNA sequencing provides an alternative that avoids some dissociation artifacts [7].

Library preparation involves reverse transcription, amplification, and the addition of sequencing adapters and barcodes. Unique molecular identifiers allow the removal of amplification bias during data analysis. The specific steps depend on the platform and the modality being measured [5].

Sequencing Strategy

Sequencing depth and read length affect the quality and cost of single-cell data. For scRNA-seq, the goal is to sequence enough reads per cell to accurately quantify gene expression without wasting resources on saturation. The optimal depth depends on the platform, the number of cells, and the biological question [12]. For single-cell genome sequencing, deeper coverage is typically needed to detect mutations with confidence.

Data Analysis Pipeline

Read Mapping and Quantification

The computational analysis of single-cell data begins with read mapping to a reference genome or transcriptome, followed by gene expression quantification [5]. This step generates a count matrix in which rows represent genes and columns represent cells. The choice of alignment and quantification tools affects downstream results, and researchers should document their pipeline carefully.

Quality Control

Quality control is essential in single-cell analysis. Low-quality cells, empty droplets, and doublets can distort downstream analyses. Standard quality control metrics include the number of genes detected per cell, the total number of reads or transcripts, and the fraction of reads mapping to mitochondrial genes [5][12]. High mitochondrial read fractions typically indicate damaged or dying cells. Doublet detection methods identify cells that were captured as pairs, which would otherwise appear as intermediate states between two cell types.

The thresholds for these metrics depend on the tissue, the dissociation protocol, and the platform. Researchers should examine the distributions of quality metrics and set thresholds based on the data instead of applying universal cutoffs [12].

Normalization

Normalization adjusts for differences in sequencing depth and capture efficiency across cells. Various methods exist, and the choice of normalization approach can affect downstream analyses [5][8]. Count modeling is an active area of research, with methods designed to account for the sparsity and overdispersion characteristic of single-cell data [8].

Feature Selection and Dimensionality Reduction

Single-cell datasets are high-dimensional, with tens of thousands of genes measured across thousands to millions of cells. Feature selection identifies the most informative genes, typically those with high variability across cells. Dimensionality reduction methods, including principal component analysis and nonlinear approaches, project the data into a lower-dimensional space that preserves the major sources of variation [5]. These reduced representations are used for clustering and visualization.

Clustering and Cell-Type Annotation

Clustering groups cells with similar expression profiles, identifying putative cell types and states. Many clustering algorithms are available, and the choice of algorithm and parameters affects the resulting clusters [5][8]. Suboptimal clustering can obscure rare cell populations or merge distinct cell types, with consequences for all downstream analyses [8].

Cell-type annotation assigns biological labels to clusters. This step can be performed manually based on marker gene expression, or with automated tools that compare cluster profiles to reference datasets [8]. Annotation reliability is a major challenge, particularly for novel or disease-associated cell states that do not match existing references [8].

Trajectory Inference

Trajectory inference methods order cells along developmental or activation continua, reconstructing the transitions between cell states [5]. These analyses can reveal the progression of differentiation, the emergence of disease-associated states, and the relationships between cell populations. Trajectory inference is sensitive to the quality of clustering and the assumptions of the algorithm.

Differential Expression and Cell-Cell Communication

Differential expression analysis identifies genes that differ between cell types, conditions, or trajectories. The choice of statistical method matters, and inadequate methods can produce false positives or miss true differences [8]. Cell-cell communication inference uses ligand-receptor co-expression to predict interactions between cell populations [8]. These analyses generate hypotheses about intercellular signaling that can be tested experimentally.

Machine Learning and Deep Learning Approaches

Machine learning methods are increasingly used throughout the single-cell analysis pipeline. In single-cell genomics, machine learning facilitates clustering, trajectory inference, and cell type identification [22]. Deep learning approaches have shown promising results in data preprocessing and downstream analysis, though they have not yet resolved all the pressing challenges of the field [26]. The high dimensionality, sparsity, and complexity of single-cell data make traditional computational approaches difficult, and deep learning offers alternatives for feature extraction and pattern recognition [26].

At a Glance

Practical Implementation Steps

Step 1: Define the Biological Question

Write a clear statement of the question the experiment will answer. Specify the expected cell types, the need for rare cell detection, and the comparisons that will be made. This statement guides decisions about modality, cell numbers, and sequencing depth [10][12].

Step 2: Select the Platform

Evaluate available platforms against the requirements of the biological question. Consider throughput, sensitivity, cost, and the types of samples that can be processed. Consult current literature and platform documentation, as the technology changes rapidly [12].

Step 3: Plan Sample Collection and Preservation

Determine whether samples can be processed immediately or require preservation. If preservation is needed, select a method validated for the sample type and the platform [15]. Document the time between collection and processing, as this affects data quality.

Step 4: Perform Quality Control at Every Stage

Assess cell viability before library preparation. Monitor the number of cells captured, the sequencing saturation, and the quality metrics after sequencing. Record all quality control decisions, including thresholds and filtering criteria [5][12].

Step 5: Document the Analysis Pipeline

Record the software versions, parameters, and reference genomes used for every analysis step. This documentation is essential for reproducibility and for interpreting results in the context of the methods used [5].

Step 6: Validate Findings

Confirm key findings with orthogonal methods. For example, validate cell-type annotations with immunostaining or flow cytometry, and validate differential expression with quantitative PCR or protein measurements. Single-cell data generate hypotheses that require independent confirmation.

Records and Measurements

Maintain detailed records of experimental and analytical decisions. The following measurements should be recorded for every single-cell experiment:

• Sample metadata, including source, collection time, preservation method, and processing time
• Cell counts and viability before and after dissociation
• Sequencing metrics, including reads per cell, genes detected per cell, and saturation
• Quality control thresholds and the number of cells removed at each step
• Software versions and parameters for every analysis step
• Clustering parameters and the number of clusters identified
• Cell-type annotation decisions and the markers used

These records support reproducibility and allow other researchers to evaluate the robustness of the findings [5][12].

Common Failure Patterns

Inadequate Sample Quality

Poor cell viability at the start of an experiment produces low-quality data that cannot be rescued by computational methods. High mitochondrial read fractions, low gene detection, and batch effects often trace back to sample handling. The solution is to optimize dissociation protocols and minimize the time between collection and processing [12][15].

Insufficient Cell Numbers

Too few cells limit the detection of rare populations and reduce statistical power for differential expression. The required number depends on the expected frequency of the rarest cell type of interest and the number of conditions being compared [10].

Batch Effects

Samples processed on different days, by different technicians, or with different reagent lots can show systematic differences unrelated to biology. Batch effects can be mistaken for biological variation. Experimental design should include batch balancing, and analysis should include batch correction methods [8].

Overclustering and Underclustering

Choosing too many clusters splits real cell types into artificial subgroups. Choosing too few merges distinct populations. Both errors propagate to downstream analyses. Cluster resolution should be evaluated with multiple methods and validated with marker genes [8].

Annotation Errors

Mislabeling clusters can lead to incorrect biological conclusions. Automated annotation tools reduce the burden of manual annotation but depend on the quality of reference data. Manual review of marker genes remains important [8].

Overinterpretation of Trajectories

Trajectory inference produces a computational ordering of cells, not a direct observation of developmental transitions. The inferred trajectories depend on algorithm assumptions and can be misleading when the underlying biology does not match those assumptions.

Limitations and Interpretation Boundaries

Single-cell genomics has inherent limitations that constrain interpretation. The technology measures molecular features at a single time point, providing a snapshot instead of a dynamic view. The act of cell isolation can alter gene expression, particularly for stress-responsive genes. Dissociation protocols can lose fragile cell types or activate transcriptional programs [12].

Sparsity is a defining feature of single-cell data. Many genes are not detected in every cell, even when they are expressed, because of the limited capture efficiency of current methods [8]. This dropout creates challenges for analysis and can be mistaken for biological variation.

The lack of universal standardization in analysis methods is a recognized limitation of the field [5]. Different pipelines can produce different results from the same data. Researchers should therefore interpret findings in the context of the specific methods used and should validate key conclusions with independent approaches.

Data integration across datasets, modalities, and species remains challenging. Assumptions in integration methods can distort biological signals, and the reliability of cell annotations across datasets is a persistent concern [8].

Data Sharing and Reproducibility

Single-cell datasets are large and complex, and their value extends beyond the original study. Data sharing enables reanalysis, meta-analysis, and the development of improved computational methods. Public repositories such as the NCBI Data Resources provide infrastructure for depositing and accessing genomic data [2]. The EMBL-EBI Training program offers educational resources for researchers learning to work with these data [1].

Funding agencies and journals increasingly require data sharing. The NIH Genomic Data Sharing Policy establishes expectations for the deposition and sharing of genomic data generated with NIH funding [3]. Researchers should be aware of the requirements that apply to their work and should plan for data deposition from the start of a project.

The FAIR Guiding Principles describe the characteristics of findable, accessible, interoperable, and reusable data [4]. Applying these principles to single-cell data involves using standard file formats, providing rich metadata, and depositing data in recognized repositories. FAIR data practices increase the impact of research and support the cumulative progress of the field.

Applications Across Research Domains

Cancer Research

Single-cell genomics has transformed cancer research by revealing intratumoral heterogeneity, clonal evolution, and the composition of the tumor microenvironment [18]. Single-cell analysis identifies rare drug-resistant populations, characterizes immune cell infiltration, and maps the spatial organization of tumors [18]. These insights inform treatment strategies and the development of immunotherapies.

Neuroscience

Single-cell and spatial transcriptomics have enabled the construction of brain atlases in mouse, human, non-human primate, and zebrafish [13]. These atlases define cellular heterogeneity across brain regions and characterize the spatial organization of neuronal circuits [13]. Spatial genomics technologies are particularly valuable for identifying region-specific molecular signatures associated with neurological disorders [13].

Immunology and Infectious Disease

Single-cell analysis characterizes immune cell states, activation trajectories, and cell-cell communication in health and disease [8]. These approaches reveal the heterogeneity of immune responses and identify cell populations associated with protection or pathology.

Developmental Biology

Single-cell genomics provides a window into developmental processes, revealing the trajectories by which progenitor cells differentiate into mature cell types [6]. These studies have identified novel intermediate states and the gene regulatory networks that control cell fate decisions.

Aging Research

Single-cell sequencing technologies have been applied to human aging research across genomics, transcriptomics, epigenomics, and proteomics [11]. These approaches have identified aging biomarkers, elucidated signaling pathways associated with aging, and discovered novel aging cell subpopulations [11]. Integration of single-cell technologies with spatial transcriptomics and mass spectrometry has expanded the scope of aging research [11].

Adipose Tissue Biology

Single-cell transcriptomics has transformed understanding of adipose tissue, revealing the diverse array of cell types and cell states that contribute to tissue plasticity [7]. These studies have provided insight into how transcriptional changes in individual cell types contribute to the capacity of adipose tissue to change in size and composition under physiological and pathophysiological conditions [7].

Forensic Science

Single-cell DNA analysis has applications in forensic science, particularly for resolving mixed DNA samples [21]. By subsampling individual cells from a mixed stain, it is possible to recover single-source DNA profiles from each donor [21]. This approach can detect low-level minor contributors that standard bulk methods miss and can resolve mixtures involving first-degree relatives [21].

Agricultural and Veterinary Science

Single-cell genomics has been applied to livestock species to understand muscle development, immune responses, and production traits [16][27]. A study of pig skeletal muscle constructed a single-cell atlas and identified cell-type-specific expression quantitative trait loci associated with meat quality and growth traits [16]. These findings provide functional context for non-coding variants implicated by genome-wide association studies and may inform genomic selection and precision genome editing in livestock [16].

Microbiology

Microbial single-cell sequencing technologies provide resolution to study microbial ecosystems, revealing cellular heterogeneity, interactions, and genetic evolution of microorganisms [14]. These approaches have been applied across microbial domains, with emerging methods for microbial spatial omics using advanced imaging techniques [14].

Emerging Technologies and Future Directions

The field of single-cell genomics continues to evolve rapidly. Emerging technologies include improved methods for multi-omics integration, spatial transcriptomics at higher resolution, and computational tools for integrating and interpreting multimodal datasets [8][9]. Deep learning methods are being developed for data preprocessing, clustering, trajectory inference, and cell-cell communication analysis [22][26].

Web-based tools are making single-cell analysis more accessible. For example, scRAPID-web provides a user-friendly interface for predicting protein-RNA interactions from single-cell transcriptomics data across eight model organisms [19]. Such tools democratize access to advanced analyses, though users should understand the underlying methods and their limitations.

The integration of single-cell genomics with other technologies, including CRISPR screening and spatial omics, is expanding the range of questions that can be addressed [9]. These integrated approaches promise to connect molecular features to function and to map cellular behavior within intact tissues.

Professional Escalation Criteria

Researchers should seek expert consultation when they encounter situations beyond their current expertise. The following situations warrant escalation to a bioinformatics specialist, a platform vendor, or a core facility:

• Unusual quality control metrics that persist after standard filtering
• Batch effects that cannot be corrected with standard methods
• Clustering results that are unstable across parameter choices
• Cell-type annotations that conflict with established biology
• Trajectory inferences that contradict known developmental relationships
• Computational resource requirements that exceed local capacity
• Questions about regulatory requirements for data sharing or human subjects research

Core facilities and vendor technical support can provide guidance on platform-specific issues. Bioinformatics specialists can advise on analysis methods and interpretation. For clinical applications, consultation with molecular pathologists and regulatory experts is appropriate.

Frequently Asked Questions

What is the difference between single-cell genomics and bulk genomics?

Bulk genomics measures average signals across millions of cells, masking heterogeneity and rare populations [26]. Single-cell genomics assigns molecular measurements to individual cells, preserving information about cell-to-cell variation [6]. This resolution enables the identification of cell types, rare populations, and developmental trajectories that bulk methods cannot resolve.

How many cells do I need for a single-cell RNA sequencing experiment?

The required number of cells depends on the biological question, the expected frequency of rare cell types, and the number of conditions being compared [10]. Experiments range from thousands to millions of cells. Researchers should consider the statistical power needed for differential expression and the likelihood of detecting rare populations when determining cell numbers.

What is the difference between single-cell and single-nucleus RNA sequencing?

Single-cell RNA sequencing analyzes intact cells, while single-nucleus RNA sequencing analyzes isolated nuclei. Single-nucleus approaches are useful for tissues that are difficult to dissociate, such as adipose tissue and brain [7]. They avoid some dissociation artifacts but capture primarily nuclear RNA, which may differ from cytoplasmic RNA.

How do I choose between different single-cell platforms?

Platform choice depends on throughput, sensitivity, cost, and sample compatibility [12]. Droplet-based platforms capture thousands to millions of cells at lower cost per cell. Other platforms offer higher sensitivity or the ability to process specific sample types. Researchers should evaluate current options against their specific biological question, as the technology changes rapidly.

What causes batch effects in single-cell experiments and how can I avoid them?

Batch effects arise from systematic differences between samples processed at different times, by different technicians, or with different reagents. They can be mistaken for biological variation. Experimental design should include batch balancing, and analysis should include batch correction methods [8]. Documenting all processing details supports the identification and correction of batch effects.

Can I combine single-cell data with spatial transcriptomics?

Yes, integration of single-cell and spatial transcriptomics is a standard approach in many studies [8][13]. Single-cell data provide high-resolution molecular profiles, while spatial data provide anatomic context [13]. Integration methods map single-cell profiles to spatial locations, enabling the characterization of tissue organization and cell-cell interactions.

What are the main challenges in analyzing single-cell data?

Key challenges include data sparsity, high dimensionality, batch effects, annotation reliability, and the lack of universal standardization in analysis methods [5][8]. Different pipelines can produce different results from the same data. Researchers should document their methods carefully and validate key findings with independent approaches.

How should I share single-cell data?

Single-cell data should be deposited in public repositories such as the NCBI Data Resources [2]. Researchers should follow applicable data sharing policies, including the NIH Genomic Data Sharing Policy for NIH-funded research [3]. Applying the FAIR Guiding Principles, including the use of standard file formats and rich metadata, increases the findability, accessibility, interoperability, and reusability of the data [4].

Related Bioinformatics Guides

• Single-Cell RNA-Seq Analysis Pipelines for Veterinary Immunology
• Foundation Models for Single-Cell Biology
• Single-Cell ATAC-Seq Bioinformatics
• Alternative Splicing Analysis from RNA-Seq Data
• Single-Cell RNA Sequencing: From Bulk to Resolution

References and Further Reading

• EMBL-EBI Training. European Bioinformatics Institute.
• NCBI Data Resources. National Center for Biotechnology Information.
• Genomic Data Sharing Policy. National Institutes of Health.
• The FAIR Guiding Principles. Scientific Data.
• Single-Cell RNA Sequencing Analysis: A Step-by-Step Overview.. Methods in molecular biology (Clifton, N.J.), 2021.
• Single-Cell Genomics.. Clinical chemistry, 2019.
• Adipose tissue at single-cell resolution.. Cell metabolism, 2023.
• A Review of Single-Cell RNA-Seq Annotation, Integration, and Cell-Cell Communication.. Cells, 2023.
• Single-cell omics: experimental workflow, data analyses and applications.. Science China. Life sciences, 2025.
• Design and Analysis of Single-Cell Sequencing Experiments.. Cell, 2015.
• Investigation of human aging at the single-cell level.. Ageing research reviews, 2024.
• Practical Considerations for Single-Cell Genomics.. Current protocols, 2022.
• Brain Spatial Genomics Atlases.. 2026.
• Microbial single-cell sequencing technologies: Principles, applications, and spatial integration.. 2026.
• Multisite Assessment of Methods for Cell Preservation Upstream of Single-Cell RNA Sequencing.. 2026.
• A scalable framework for single-cell eQTL mapping uncovers genetic regulators of meat production traits in pigs.. 2026.
• Corrigendum to "Cross-species single-cell and spatial transcriptomic mapping reveals EFNA1-EPHA4-mediated stem-like epithelial-macrophage crosstalk driving colorectal cancer progression" [Cancer Letters 649 (2026) 218495].. 2026.
• Thematic Shifts in Early-High-Impact Cancer Genomics and Diagnostics Research: A Bibliometric and Semantic Analysis. 2026.
• scRAPID-web: a web server for predicting protein-RNA interactions from single-cell transcriptomics.. 2026.
• Advances in Genetics and Genomics of Animal Breeding: Call for Papers. 2026.
• Single cell genomics applications in forensic science: Current state and future directions. iScience, 2023.
• Machine Learning in Genomics: Applications in Whole Genome Sequencing, Whole Exome Sequencing, Single-Cell Genomics, and Spatial Transcriptomics. Medinformatics, 2024.
• Preparation of single cell suspensions from human intestinal biopsies for single cell genomics applications v1. protocols.io, 2020.
• Preparation of single cell suspensions from human intestinal biopsies for single cell genomics applications v2. protocols.io, 2020.
• Preparation of single cell suspensions from human intestinal biopsies for single cell genomics applications v3. protocols.io, 2020.
• Deep learning applications in single-cell genomics and transcriptomics data analysis.. Biomedicine & pharmacotherapy = Biomedecine & pharmacotherapie, 2023.
• Single-Cell Genomics in Veterinary Science: Methods and Applications. Bioinformatics in Veterinary Science Vetinformatics, 2025.
• Single-cell genomics: An overview. Frontiers in Biology, 2013.
• Human gut evolution: insights from stem cell models and single-cell genomics. Current Opinion in Genetics and Development, 2025.

This article is educational and does not replace validated analysis plans, institutional policy, clinical interpretation, or specialist review.