Revealing the Critical Regulators of Cell Identity in the Mouse Cell Atlas

Abstract

SUMMARYRecent progress in single-cell technologies has enabled the identification of all major cell types in mouse. However, for most cell types, the regulatory mechanism underlying their identity remains poorly understood. By computational analysis of the recently published mouse cell atlas data, we have identified 202 regulons whose activities are highly variable across different cell types, and more importantly, predicted a small set of essential regulators for each major cell type in mouse. Systematic validation by automated literature and data mining provides strong additional support for our predictions. Thus, these predictions serve as a valuable resource that would be useful for the broad biological community. Finally, we have built a user-friendly, interactive web portal to enable users to navigate this mouse cell network atlas.

Mammalian gene expression variability is explained by underlying cell state

“…Gene expression variability is ubiquitous in all biological systems. In multicellular organisms, heterogeneity between different cell types and states confers specialized function giving rise to complexity in whole‐system behavior (Raj & van Oudenaarden, ; Eldar & Elowitz, ; Symmons & Raj, ; Suo et al , ; Tabula Muris Consortium et al , ). Similarly, single‐cell organisms and viruses were shown to utilize heterogeneity at the population level to create diverse phenotypes, such as bet‐hedging strategies in changing environments (Veening et al , ; Vega & Gore, ; Rouzine et al , ).…”

Predicting gene regulatory networks from cell atlases

“…The high degree of regulon overlap between the three atlases, in spite of technical differences, highlights that single-cell regulatory state is predominantly governed by core set regulators and their activities within individual cells. A recent study also applied SCENIC, but only for MCA data using only the author-assigned cell type labels (21).…”

“…To highlight the regulon crosstalk and regulation across integrated cell atlas, we performed regulon-to-regulon correlation using Connection Specificity Index (CSI) Supplemental Data 1 (21,33). The CSI is a context dependent graph metric that ranks the regulon significance based on similarity and specificity of interaction partners, thereby mitigating the effects of non-specific interactions.…”

“…Although direct comparison between the two approaches is tricky because of underlying scoring methods, we plot the regulon correlation between VIPER enrichment scores and mean RAS (NES versus mean AUC; Fig S22A). However, the correlation is significantly improved when considering the VIPER enrichment scores (42) with regulon specificity scores (21), indicating specific cell type enrichment (Fig S22B).…”

“…For regulatory and mechanistic insights beyond cell type survey across the three atlases, we have to extend analysis beyond comparison of gene expression patterns. The computational inference of TFs and their regulated gene sets (regulons) provides an avenue to extract the regulatory crosstalk from single-cell expression data (9,10,21,22). Here, we set out to comprehensively reconstruct GRNs from single-cell atlases and address the following questions: (i) Which TFs, master regulators, and co-factors (i.e., regulons) govern tissue and cell types?…”