Overview of Chromatin Accessibility Analysis “Strategies and Tools”

Team-Meetings. May 21st
Summary:
scATAC-seq enables the discovery of regulatory landscapes by profiling chromatin accessibility at single-cell resolution, revealing cell-specific regulatory mechanisms.
Multi-omic approaches, which integrate chromatin accessibility and gene expression from the same nucleus, provide enhanced biological insights. In contrast, separate analyses offer greater flexibility in experimental design and tool selection.
Key tools like Seurat, Signac, MACS2, Cicero, and chromVAR form a powerful framework for data processing, clustering, enhancer-gene linkage, and transcription factor motif analysis.
Case studies in brain and immune development demonstrate conserved regulatory logic, cell-type-specific cis-elements, and disease-associated regulatory variants.
The field is rapidly evolving, making it essential to explore new integrative tools and functional assays to map the dynamic regulatory genome.

Transcript

1. Overview of Chromatin Accessibility Analysis “Strategies and Tools” Cynthia Soto Cardinault May 21, 2025 Team meetings
3. States of chromatin Chromatin can exist in two main states: euchromatin and heterochromatin Euchromatin is relatively open and accessible, often containing actively transcribed genes, while heterochromatin is highly condensed and less accessible, typically housing inactive genes Histone modifications or epigenetic features that are associated with active regulatory elements Active marks are often identified via ChIP-seq 3
4. ATAC-Seq is based on transposase-mediated insertion of NGS adaptors into open chromatin regions. 4
6. Regulatory Element Description Promoter Located near the TSS; contains a TATA box; recruits RNA Pol II for transcription initiation. Enhancer Distal elements that boost gene expression by recruiting activators; interact with promoters via DNA looping. Locus Control Region (LCR) Long-range cis-regulatory hub controlling entire gene loci; ensures tissue-specific, position-independent expression. 6 Elements of gene regulation
7. Common regulatory elements Regulatory Element Description Expected Distance Promoter Located near the TSS; contains a TATA box; recruits RNA Pol II for transcription initiation. ≤ 2 kb from the TSS Enhancer Distal elements that boost gene expression by recruiting activators; interact with promoters via DNA looping. 10–100 kb, up to 500 kb+ Locus Control Region (LCR) Long-range cis-regulatory hub controlling entire gene loci; ensures tissue-specific, position-independent expression. Tens of kb to >2 Mb 7
9. Roadmap of a typical ATAC-seq analysis Yan, F., Powell, D.R., Curtis, D.J. et al (2020) Zhu K, Bendl J, et al 2022 Habenula project Yan, F., Powell, D.R., Curtis, D.J. et al (2020) 9
10. Yan, F., Powell, D.R., Curtis, D.J. et al (2020) 10
11. 11 Strategies and tools used for chromatin accessibility analysis
12. Study Overview • Created a multi-omic atlas of the human cerebral cortex across six developmental stages (fetal to adult). • Simultaneous profiling ATAC-seq and RNA-seq in the same nucleus using single-nucleus multiome sequencing. 12
13. Methodology Overview 10x Genomics Single Cell Multiome ATAC + Gene Expression multi-omic profiling • Samples: 12 human neocortex samples across 6 developmental stages • Captured nuclei: 53k → 45.5 QCed cells Cell Ranger ARC v1.0.0: data preprocessing • Read alignment (to GRCh38) • Barcode assignment & filtering • ATAC peak calling (initial) • RNA/ATAC molecule quantification per nucleus 13
14. Quality Control Metrics: Seurat and Signac • RNA-seq: Filtered cells with low gene counts, high mitochondrial gene expression, or other quality issues. • ATAC-seq: Assessed fragment size distribution (NFR), transcription start site (TSS) enrichment, and other standard QC metrics. QC applied on ATAC: • NS<3 • Minimum Number of Fragments>1,000 • TSS Enrichment Score>2 • A FRiP score>0.2: below 0.2 led to exclusion this metric reflects the proportion of reads that fall within called peak regions Gene associations are enriched in the vicinity of TSSs, and the correlations decayed exponentially with distance. (Kaiyi Zhu, et al, 2023) 14
15. ENCODEUpdated July 2020 Yan et al (2020) Typical peak annotation More than half of the peaks fall into enhancer regions (distal intergenic and intronic regions), And only around 25% of the peaks are in promoter regions. TSS: transcription start site 15
16. MACS2: peak calling • Peak calling for aggregated scATAC-seq data • Parameters optimized for ATAC-seq (e.g., --nomodel, --shift, --extsize) Seurat & Signac: data integration & clustering • WNN: RNA (SCTransform) + ATAC (LSI) • Adjusted Rand Index (ARI) used to validate modality concordance Why do the authors used MACS2 instead of Signac for pseudo-bulk peaks? - MAC2 is a well-established algorithm optimized for bulk ATAC-seq and ChIP-seq data. - especially helpful when defining consensus regulatory elements across cell types or developmental stages. - Signac's peak calling (via CallPeaks()) relies on MACS2 under the hood or a more basic windowing strategy when used on single cells. 16
17. Pseudotime & Enhancer-Gene Linkage • Monocle 3: pseudotime inference • Cicero: Correlation-based enhancer-gene mapping 17
18. Transcription Factor Motif Analysis • chromVAR: Quantify cell-to-cell variability in TF motif • JASPAR2020 Motif Database:It was used to link peaks to specific TF motifs, identifying key regulators of neural development. GWAS Enrichment: mapped accessible regions to SNPs from neuropsychiatric GWAS. LDSC regression: Heritability enrichment in ATAC peaks: fetal regulatory elements enriched for schizophrenia risk variants. chromVAR is designed for single cell or bulk ATAC-seq, and works well with multiome datasets integrated using Signac, which the authors used. 18
19. Chopp LB, et al. 2020 (19k QCed) Cell Ranger (scRNA-seq) v2.2.0 Cell Ranger ATAC (scATAC-seq) v1.1.0 with mm10 for mouse and GRCh38-3.0.0 for human Seurat v3.1(std log-normalization) Signac v0.2.1 and v1.0.0 (LSI) Data integration: • Monocle3 • MAC2 v2.2.5 • HOMER v4.10: Motif Enrichment Analysis Cellranger-arc (v.1.0.0) Seurat v4.0.13 (SCTransform) Signac v1.1.0 (LSI) WNN (SLM) Monocle3: Trajectory alignment for scRNA-seq pseudotime. MACS2: identify peaks Cicero: Correlated peaks chromVAR: TF enrichement Zhu K, et al 2022 (41k QCed) Morabito S, et al. 2022 (191k QCed) Cell Ranger (scRNA-seq) v3.0.2 Cell Ranger ATAC (scATAC-seq) 1.1.0 Seurat v3.1.4 (SCTransform) Signac v0.2.5 Data integration: • Monocle 3 v0.2.3.0 • MACS2 v2.1.2 • Cicero v1.3.4.10 • chromVAR v1.8.0 19
20. Some nuances to consider How Much Abundance Is Needed to Detect cis-Regulatory Elements? About 1000 nuclei were sufficient to identify over 80% of the accessible regions in a cell type, consistent with our previous finding Li YE,et al (2023) Are Cell Barcodes Accurately Called in Cell Ranger ARC? The cell-calling algorithm uses the paired data to distinguish cells from non-cell background. But, if a barcode fails any of the three filtering criteria, the data is masked from the total set of barcodes prior to cell calling. 10x Genomics (2025) Are Chromatin Accessibility Peaks Proximal to the TSS? Only 22% of the peak-gene links occur between an ATAC-seq peak and the nearest gene,indicating that most predicted regulatory interactions skip at least one gene along the linear genome Kaiyi Zhu, et al (2023) 20
21. Chang Su, Dongsoo Lee, Peng Jin, Jingfei Zhang (2024) Zhang, K., Zemke, N. R., Armand, E. J. & Ren, B. (2024) Lu RJ, Liu YT, Huang CW, Yen MR, Lin CY, Chen PY. (2021) Other tools ? 21
22. Takeaway Summary • scATAC-seq enables the discovery of regulatory landscapes by profiling chromatin accessibility at single-cell resolution. • Multi-omic approaches integrate chromatin accessibility and gene expression from the same nucleus, enhancing biological resolution, while separate analyses offer greater flexibility in experimental design and tool selection. • Key tools like Seurat, Signac, MACS2, Cicero, and chromVAR provide a robust framework for data processing, clustering, enhancer-gene linkage, and TF motif analysis. • Case studies across brain and immune development highlight conserved regulatory logic, cell-type-specific cis-elements, and disease-associated regulatory variants. • The field is rapidly evolving—keep exploring new integrative tools and functional assays to map the dynamic regulatory genome. 22