Zubair Khalid

Virologist/Molecular Biologist | Veterinarian | Bioinformatician

Conventional & Molecular Virology • Vaccine Development • Computational Biology

Dr. Zubair Khalid is a veterinarian and virologist specializing in conventional and molecular virology, vaccine development, and computational biology. Dedicated to advancing animal health through innovative research and multi-omics approaches.

Dr. Zubair Khalid - Veterinarian, Virologist, and Vaccine Development Researcher specializing in Computational Biology, Multi-omics, Animal Health, and Infectious Disease Research

Category: Guides

Single Cell ATAC Seq: A Practical Guide for Epigenomic Researchers

Single cell ATAC seq (Assay for Transposase Accessible Chromatin with sequencing at single cell resolution) maps chromatin accessibility genome wide in individual cells, revealing regulatory landscapes and cell state heterogeneity. This guide is written for bench scientists, bioinformaticians, and principal investigators who are planning their first single cell ATAC seq experiment or interpreting published data. For a reliable technical foundation, consult the free resources at the NCBI Bookshelf. The methods described here are informed by established protocols and recent discoveries, including work that traced B lineage bifurcation using epigenetic signatures [6] and a cost effective high throughput approach for measuring chromatin accessibility in single nuclei [11].

At a Glance: Single Cell ATAC Seq Essentials

Aspect Key Information
What it measures Genome wide chromatin accessibility per single nucleus (open chromatin regions).
Typical input 1,000 to 50,000 nuclei (fresh or frozen) depending on platform.
Library prep time 1 to 2 days after nuclei isolation.
Sequencing depth 20,000 to 50,000 reads per cell recommended, lower depths reduce peak detection.
Output Fragments mapped to genome, peak calls per cell, cell by peak matrix.
Main uses Cell type identification, transcription factor footprinting, regulatory element discovery, integration with single cell RNA seq.

Core Concepts and Biological Context

Single cell ATAC seq uses a hyperactive Tn5 transposase that preferentially inserts sequencing adapters into open (accessible) chromatin. After amplification and sequencing, read counts at genomic regions reflect chromatin accessibility per cell. This differs from bulk ATAC seq which averages across many cells. The technique captures regulatory heterogeneity, such as the early lineage bifurcation in human B cells that was identified through epigenetic marks [6]. Recent developments also enable simultaneous profiling of histone modifications and transcription factor binding at single cell resolution [7]. The EMBL EBI Training provides excellent modules on the biological principles of chromatin accessibility assays.

One key concept is that open chromatin correlates with active regulatory elements. However, accessibility alone does not confirm transcription factor binding or gene expression. Multiomic approaches that integrate ATAC seq with RNA seq or protein data can resolve these connections, as demonstrated in studies of CAR T cell efficacy [10] and EMT tumor states [8].

Decision Points for Experimental Design

Choose your platform based on throughput and cost. Droplet based methods (10x Genomics, Drop seq) capture thousands to tens of thousands of cells per run, while plate based methods (Smart seq3 like adaptations) capture hundreds but with deeper coverage per cell. For most projects, droplet based single cell ATAC seq is the standard.

Consider whether you need fresh tissue or can use frozen nuclei. Protocols for frozen samples are available but may reduce sensitivity. The method described in [11] offers a simple, cost effective, high throughput option that works with both fresh and frozen nuclei.

Another decision is whether to run paired end or single end sequencing. Paired end is strongly recommended because it improves fragment length estimation and mapping accuracy. The Galaxy Training Network includes a tutorial on single cell ATAC seq analysis that covers these choices.

Finally, plan for integration with other modalities. If you aim to link accessibility to gene expression, consider a multiomic kit or computational integration after separate assays. The conserved co regulation of pyrophosphate homeostasis genes [9] was uncovered by combining ATAC seq with RNA seq in both mice and humans.

Practical Workflow or Implementation Steps

  1. Nuclei isolation and tagmentation
    Isolate nuclei from cells or tissue. Resuspend in a buffer that maintains intact nuclei. Add Tn5 transposase loaded with sequencing adapters. Incubate for 30 minutes at 37 degrees Celsius. Stop the reaction with a clean up step.

  2. Single cell partitioning
    For droplet methods, load nuclei onto a microfluidic chip with barcoded beads. Each bead carries a unique barcode that tags all fragments from a single nucleus. For plate methods, sort single nuclei into wells.

  3. Library amplification
    Perform gap repair and PCR amplification using indexed primers. Barcodes and sample indices are added during this step. Clean up with SPRI beads.

  4. Sequencing
    Sequence on Illumina platforms. Recommended depth: 20,000 to 50,000 read pairs per cell. Quality control metrics include fraction of reads in peaks (FRIP), transcription start site enrichment, and number of unique fragments per cell.

  5. Primary data processing
    Demultiplex samples, align reads to reference genome using a dedicated aligner (e.g., BWA, STAR). Remove duplicates, filter for mitochondrial reads, and create a fragment file. The Bioconductor package Signac is widely used for downstream analysis.

  6. Peak calling and cell by peak matrix
    Call peaks using MACS2 on pseudo bulk data or using a single cell aware tool. Generate a matrix counting fragments per peak per cell. Filter low quality cells and peaks.

  7. Downstream analysis
    Perform dimensionality reduction (LSI, UMAP), clustering, differential accessibility testing, motif enrichment, and integration with transcriptomic data. The NCBI Sequence Read Archive [5] hosts many public single cell ATAC seq datasets that you can use for validation.

Quality Checks and Metrics

After alignment, check the number of unique nuclear fragments per cell. A typical high quality cell has 3,000 to 10,000 unique fragments. Remove cells with fewer than 1,000 fragments or more than 20% mitochondrial reads.

Compute the fraction of reads in peaks (FRIP). A FRIP above 0.3 is good, above 0.5 is excellent. Also evaluate TSS enrichment: a sharp peak at the transcription start site indicates good signal.

Examine the fragment length distribution. A periodic pattern with a peak at about 147 base pairs (nucleosome protected) and a shoulder at shorter lengths (open chromatin) confirms successful tagmentation. The NCBI Bookshelf provides detailed quality control guidelines in its epigenomics chapter.

Common Mistakes and How to Avoid Them

Mistake 1: Including too many low quality cells.
Some users keep all cells with any fragments. Instead, apply strict filters. Low quality cells add noise to clustering and differential analysis.

Mistake 2: Ignoring batch effects.
If you process samples on different days or chips, use integration methods such as Harmony or Seurat’s reciprocal PCA. Failure to correct batch effects can create false clusters.

Mistake 3: Using single end sequencing.
Single end reads lose fragment length information and reduce mapping accuracy. Always perform paired end sequencing for single cell ATAC seq.

Mistake 4: Overinterpreting lack of accessibility.
Closed chromatin does not always mean the locus is permanently silenced. It may be induced later during development. Do not conclude a gene is inactive just because its promoter is inaccessible in your dataset.

Limits of Interpretation and Uncertainty

Single cell ATAC seq identifies open chromatin but cannot directly measure transcription factor binding or chromatin conformation. Footprinting analysis can infer binding sites, but requires high coverage (over 100,000 fragments per cell) that is rarely achieved in typical experiments. Integration with other assays like CUT&Tag or RNA seq can strengthen conclusions.

Another limitation is that ATAC seq does not capture all regulatory elements. Some enhancers are only active in specific conditions or cell states that may be underrepresented. The method is also biased toward open regions, very tight heterochromatin is poorly targeted.

The cost and technical complexity remain high for large scale studies. Public data from the NCBI Sequence Read Archive can supplement your own experiments, but careful normalization is needed to combine datasets from different platforms.

Interpretation of cell type specific accessibility changes requires robust clustering and often independent validation. For example, the transcriptional co regulation described in [9] was validated by separate knockout experiments. Single cell ATAC seq alone should be considered a hypothesis generating tool.

Frequently Asked Questions

Q1: How many cells do I need for a typical single cell ATAC seq experiment?
A: For most cell type discovery applications, aim for at least 5,000 high quality cells per sample. Fewer cells may still work if you expect large, discrete populations. For transcription factor footprinting, you need 10,000 or more cells per sample to achieve sufficient depth.

Q2: Can I use frozen tissue for single cell ATAC seq?
A: Yes, protocols have been optimized for frozen tissue. The method described in [11] uses frozen nuclei and yields data comparable to fresh tissue. However, expect lower nuclei recovery and slightly fewer fragments per cell compared to fresh samples.

Q3: How do I integrate single cell ATAC seq with single cell RNA seq data?
A: You can use computational methods like Seurat’s weighted nearest neighbor analysis, or you can perform a multiomic assay that simultaneously captures RNA and ATAC from the same cell. Both approaches are supported by the tools in Bioconductor.

Q4: What does a low fraction of reads in peaks (FRIP) mean?
A: A FRIP under 0.2 often indicates poor tagmentation efficiency, excessive background, or low cell quality. Check your nuclei isolation protocol and adjust tagmentation time. Increasing sequencing depth may also help, but the root cause is usually technical.

References and Further Reading

Related Articles