A comparative atlas of single-cell chromatin accessibility in the human brain

Hi there,
In this meeting, I presented a brief talk on the paper "A Comparative Atlas of Single-Cell Chromatin Accessibility in the Human Brain." The authors explore how recent advances in single-cell transcriptomics have illuminated the diversity of neuronal and glial cell types, while also addressing the still unclear regulatory mechanisms that govern cell identity and function.
In the study, authors used snATAC-seq to analyze open chromatin landscapes in 1.1 million cells from 42 brain regions across three adults. They identified 107 distinct cell types and mapped 544,735 candidate cis-regulatory DNA elements (cCREs) in the human genome. Additionally, they predicted disease-relevant cell types for 19 neuropsychiatric disorders, revealing strong links between specific brain cell types and conditions like schizophrenia, bipolar disorder, Alzheimer’s disease, and major depression.

Link to paper: DOI: 10.1126/science.adf7044

Cynthia Cardinault

Transcript

1. A comparative atlas of single-cell chromatin accessibility in the human brain Cynthia S Cardinault Aug 14, 2024
2. 1Department of Cellular and Molecular Medicine, University of California, San Diego, San Diego, CA 92093, USA 2Center for Epigenomics, University of California San Diego, School of Medicine, La Jolla, CA 92093, USA. 3The Salk Institute for Biological Studies, La Jolla, CA 92037, USA 4Department of Computer Science and Engineering, University of California San Diego, CA 92093, USA 5Division of Molecular Neurobiology, Department of Medical Biochemistry and Biophysics, Karolinska Institute, 171 77 Stockholm, Sweden 6Allen Institute for Brain Science, Seattle, WA 98109, USA 7Department of Laboratory Medicine and Pathology, University of Washington, Seattle, WA 98104, USA 8Howard Hughes Medical Institute, The Salk Institute for Biological Studies, La Jolla, CA 92037, USA
4. Introduction Using snATAC-seq analysis authors explore open chromatin landscapes across 1.1 million cells in 42 brain regions from three adults. 107 distinct cell types and their specific utilization of 544,735 candidate cis-regulatory DNA elements (cCREs) in the human genome. Nearly 1/3 of the cCREs demonstrate conservation and chromatin accessibility in the mouse brain cells. Predicted disease relevant cell types for 19 neuropsychiatric disorders. Strong links between specific brain cell types and neuropsychiatric disorders including schizophrenia, bipolar disorder, Alzheimer’s disease, and major depression Develop deep LM to predict regulatory roles of non-coding risk variants in these disorders. Interactive web atlas to disseminate this resource (cis-element ATLAS [CATLAS]; http://catlas.org ).
5. Fig 1. Single-cell analysis of chromatin accessibility in the human brain Clustering in three major classes: • Class I enriched for glutamatergic (vGlut+, putatively excitatory) neurons (11.8%) • Class II enriched for GABAergic (GABA+, putatively inhibitory) neurons (6.8%) and • Class III enriched for non-neuronal cells (81.4%).
6. Clustering the three major classes into: • 14 subclasses of vGlut+ neurons, 2 subclasses of granule cell types, 1 sub-class of cholinergic neurons, 4 sub-classes of dopaminergic neurons, 2 sub-classes of thalamic and midbrain derived neurons, • 11 subclasses of cortical GABA+ neurons, and • 8 sub-classes of non-neuronal cells Each subclass annotated match at least 3 marker genes of known brain cell types, together with the brain region where the cells reside.
7. SI and SJ. Doublet detection Removed 156,614 snATAC-seq profiles that likely resulted from potential barcode collision or doublets. (SI) Box plot shows the fraction of potential barcode collisions detected in snATAC-seq libraries using a modified version of Scrublet. (SJ) Number of nuclei retained after each step of quality control. Run Scrublet on individual samples for accurate doublet detection. Ensure the doublet score threshold effectively separates peaks and adjust if needed. Visualize predictions in 2-D embeddings; if doublets don’t cluster together, tweak the threshold or pre-processing parameters. Data reliability was confirmed by nucleosome-like periodicity and TSS A total of 1,134,360 nuclei were retained.
8. DoubletFinder (2019) is implemented to interface with Seurat. DoubletDecon (2018) is Deconvoluting doublets from single-cell RNA-sequencing data ans is also available for Seurat. DoubletDetection (2020) is a python implementation.
9. Vasoactive intestinal peptide-expressing (VIP) neurons were divided into multiple cell types with distinct chromatin accessibility at multiple gene loci. Most neuronal cell types and some glial cell types were distributed in the human brain in a non-uniform fashion (Fig. 1J). Bergmann glia (ACBGM), also called Golgi epithelial cells, were specialized. Fig 1.
10. S8. Peak calling and filtering pipeline Identified the open chromatin and cCREs in each of the 107 brain cell types. Aggregated the chromatin accessibility profiles Identified the open chromatin regions with MACS2 Filtered the resulting accessible chromatin regions based on whether they were called in at least two donors, or pseudo-replicates They used “score per million” (SPM) to correct this bias (Fig. S8C). About 1000 nuclei were sufficient to identify over 80% of the accessible regions in a cell type, consistent with our previous finding.
11. Several lines of evidence support the authenticity of the identified cCREs. First, both proximal and distal cCREs showed higher sequence conservation than random genomic regions with similar GC content (Fig. 2B). Plotted the median levels of chromatin accessibility against the maximum variation for each element (Fig. 2D) To characterize the cell type specificity of the cCREs more explicitly, they used non-negative matrix factorization to group them into 37 modules, with elements in each module sharing similar cell type specificity profiles. These cCREs together made up 8.8% of the human genome. Of these, 95.3% were located at least 2 kbp away from annotated promoter regions of protein-coding and lncRNA genes (Fig. 2A) Fig 2. Identification and characterization of candidate CREs (cCREs) across human brain cell types
12. (F) Schematic of the strategy used to identify cCREs that are positively correlated with transcription of target genes. This analysis revealed a large group of 5,113 putative enhancers that were linked to 4,775 genes more strongly expressed across all neuronal cell clusters than in non-neuronal cell types (module M1) (Fig. 2H) Fig 2.
13. Fig 3. Regional specificity of glial and neuronal cell cCREs The UMAP embeddings in the brain regional spaces showed a gradient among cell types of OGCs, OPCs, MGCs, and ASCTs (Fig. 3A–N) hypothesized that these gradients may reflect heterogeneity in cCRE usage in these glial cells across brain regions. A large number of cCREs displayed highly variable chromatin accessibility across the brain regions (Fig. 3C, G, K, and O) 55K variable cCREs made up 40.1% of total cCREs identified in OGC, 33.0% of total in OPCs, …
14. Conclusion Analyzing 1.1 million cells from 42 brain regions in three neurotypical adult subjects, they identified 107 distinct brain cell types. Characterized the chromatin accessibility status of 544k putative transcriptional regulatory elements within these cell types. Identified sequence conservation and chromatin accessibility in mouse brain cells, indicating their functional significance, and association with numerous regulatory elements with potential target genes and predicted cell types relevant to 19 neuropsychiatric traits and disorders. Developed ML models to forecast the regulatory function of disease risk variants. The atlas is an interactive web portal called CATLAS.
15. Thank you!