Chromatin Potential Identified by Shared Single-Cell Profiling of RNA and Chromatin

Abstract

Cell differentiation and function are regulated across multiple layers of gene regulation, including the modulation of gene expression by changes in chromatin accessibility. However, differentiation is an asynchronous process precluding a temporal understanding of the regulatory events leading to cell fate commitment. Here, we developed SHARE-seq, a highly scalable approach for measurement of chromatin accessibility and gene expression within the same single cell. Using 34,774 joint profiles from mouse skin, we develop a computational strategy to identify cis-regulatory interactions and define Domains of Regulatory Chromatin (DORCs), which significantly overlap with super-enhancers. We show that during lineage commitment, chromatin accessibility at DORCs precedes gene expression, suggesting changes in chromatin accessibility may prime cells for lineage commitment. We therefore develop a computational strategy (chromatin potential) to quantify chromatin lineage-priming and predict cell fate outcomes. Together, SHARE-seq provides an extensible platform to study regulatory circuitry across diverse cells within tissues.

A generalization of t-SNE and UMAP to single-cell multimodal omics

“…Not surprisingly, projecting RNA and protein velocities into the joint embedding of both modalities yielded less noisy acceleration landscapes compare to embeddings of mRNA measurements alone. We therefore anticipate that JVis will aid in the meaningful visual interpretation of data generated by emerging multimodal omics technologies such as CITE-seq [19] and SHARE-seq [13], the latter allowing to combine RNA velocity with chromatin potential.…”

Chromatin remodeling in peripheral blood cells reflects COVID-19 symptom severity

“…Domains of regulatory chromatin containing transcription factor motifs become accessible prior to downstream gene expression (16) . To test whether chromatin changes can better distinguish disease severity, differential chromatin accessibility analysis was applied to the earliest seronegative samples from COVID-19 subjects compared to healthy controls.…”

A Picture Worth a Thousand Molecules—Integrative Technologies for Mapping Subcellular Molecular Organization and Plasticity in Developing Circuits

“…As a future development, deep learning shows exciting promise for linking multiomic and spatial data across biological scales and formats (Bersanelli et al, 2016;Haas et al, 2017;Mirza et al, 2019;Nguyen and Wang, 2020). By registering genome-wide single-cell sequencing data to sparse spatial transcriptomic reference frames, deep learning computer vision methods predict spatial expression patterns with increased coverage, error reduction, and multiomic integration (Biancalani et al, 2020;Ma et al, 2020). In principle, similar computational strategies could integrate target-specific proteomic datasets with mass spectrometry imaging (MSI; Xu and Li, 2019) or multi-round protein imaging for spatial proteomics.…”