The circadian clock modulates human physiology. However, the organization of tissue-specific gene expression rhythms and how these depend on age and sex is not defined in humans. We combined data from the Genotype-Tissue Expression (GTEx) project with an algorithm that assigns circadian phases to 914 donors, by integrating temporal information from multiple tissues in each individual, to identify messenger RNA (mRNA) rhythms in 46 tissues. Clock transcripts showed conserved timing relationships and tight synchrony across the body. mRNA rhythms varied in breadth, covering global and tissue-specific functions, including metabolic pathways and systemic responses. The clock structure was conserved across sexes and age groups. However, overall gene expression rhythms were highly sex-dimorphic and more sustained in females. Rhythmic programs generally dampened with age across the body.
Across biological systems, cells undergo coordinated changes in gene expression, resulting in transcriptome dynamics that unfold within a low-dimensional manifold. While low-dimensional dynamics can be extracted using RNA velocity, these algorithms can be fragile and rely on heuristics lacking statistical control. Moreover, the estimated vector field is not dynamically consistent with the traversed gene expression manifold. To address these challenges, we introduce a Bayesian model of RNA velocity that couples velocity field and manifold estimation in a reformulated, unified framework, identifying the parameters of an explicit dynamical system. Focusing on the cell cycle, we implement VeloCycle to study gene regulation dynamics on one-dimensional periodic manifolds and validate its ability to infer cell cycle periods using live imaging. We also apply VeloCycle to reveal speed differences in regionally defined progenitors and Perturb-seq gene knockdowns. Overall, VeloCycle expands the single-cell RNA sequencing analysis toolkit with a modular and statistically consistent RNA velocity inference framework.
The ability of organisms to keep track of external time, by means of the circadian clock interacting with the environment, is essential for health. The focus of this review is recent methods to detect the internal circadian time of an omics sample. Before reaching our main topic, we introduce the circadian clock, its hierarchical structure, and its main functions; we will also explain the notion of internal time, or circadian phase, and how it differs from the geophysical time. We then focus on the role played by the clock in the maintenance of human heath, in particular in the context of cancer. Thereafter, we analyze an important methodological question: how to infer the circadian phase of unlabeled omics snapshot measurements. Answering this question could both significantly increase our understanding of the circadian clock and allow the use of this knowledge in biomedical applications. We review existing methods, focusing on the more recent ones, following a historical trajectory. We explain the basic concepts underlying the methods, as well as some crucial technical aspects of each. We conclude by reporting how some of these methods have, more or less effectively, enabled furthering our understanding of the clock and given insights regarding potential biomedical applications.
Across a range of biological processes, cells undergo coordinated changes in gene expression, resulting in transcriptome dynamics that unfold within a low-dimensional manifold. Single-cell RNA-sequencing (scRNA-seq) only measures temporal snapshots of gene expression. However, information on the underlying low-dimensional dynamics can be extracted using RNA velocity, which models unspliced and spliced RNA abundances to estimate the rate of change of gene expression. Available RNA velocity algorithms can be fragile and rely on heuristics that lack statistical control. Moreover, the estimated vector field is not dynamically consistent with the traversed gene expression manifold. Here, we develop a generative model of RNA velocity and a Bayesian inference approach that solves these problems. Our model couples velocity field and manifold estimation in a reformulated, unified framework, so as to coherently identify the parameters of an autonomous dynamical system. Focusing on the cell cycle, we implemented VeloCycle to study gene regulation dynamics on one-dimensional periodic manifolds and validated using live-imaging its ability to infer actual cell cycle periods. We benchmarked RNA velocity inference with sensitivity analyses and demonstrated one- and multiple-sample testing. We also conducted Markov chain Monte Carlo inference on the model, uncovering key relationships between gene-specific kinetics and our gene-independent velocity estimate. Finally, we applied VeloCycle to in vivo samples and in vitro genome-wide Perturb-seq, revealing regionally-defined proliferation modes in neural progenitors and the effect of gene knockdowns on cell cycle speed. Ultimately, VeloCycle expands the scRNA-seq analysis toolkit with a modular and statistically rigorous RNA velocity inference framework.
This prospective, single-center, case-control study investigated circulating autoantibodies (AAbs) targeting G protein-coupled receptors (GPCRs) in Long COVID (LC) patients to identify potential diagnostic biomarkers and therapeutic targets. Fifteen participants were enrolled at the LC clinic in Rome: eleven with severe LC—defined as >4 persistent symptoms (fatigue, cognitive impairment, poor exercise tolerance, dyspnea, arthralgia, or dysautonomic manifestations) >3 months post-infection—and four asymptomatic post-COVID (APC) individuals. Fatigue was assessed using the Fatigue Assessment Scale (FAS ≥ 22; severe ≥ 35). Auto-Abs against AT1R, endothelin receptor A, adrenergic (α1, α2, β1, β2), and muscarinic (M1–M5) receptors were quantified, along with blood cortisol and ACTH levels. SARS-CoV-2-specific T-cell responses to Spike and Nucleocapsid proteins were evaluated by ELISpot assay. In our small cohort, LC patients were younger, had fewer comorbidities (p = 0.03), fewer vaccine doses (p = 0.03), and higher FAS scores (33 vs. 12; p = 0.001). Mean GPCR AAbs levels were higher in LC than in APC (8.88 vs. 5.45 Units/mL; p = 0.17), indicating a coherent autoimmune signature in LC that correlates with symptom development. Morning cortisol was lower in LC (12.7 vs. 17 mg/dL; p = 0.01), and T-cell responses tended to be weaker. These findings suggest GPCR AAbs may serve as biomarkers and therapeutic targets for a subset of patients, guiding diagnosis and treatments with IV immunoglobulin or immunoadsorption.
The circadian clock modulates most of human physiology. However, the organization of tissue-specific gene expression rhythms is still poorly known, as well as age and sex dependencies. We leveraged the Genotype-Tissue Expression project (GTEx) using a novel algorithm to assign a unique circadian phase to 914 individuals and transfer time information from stronger to weaker clocks. These donor internal phases allowed us to identify and compare programs of rhythmic gene expression in 46 tissues. Clock transcripts showed highly conserved phase and amplitude relationships across tissues, and were tightly synchronized across the body. Tissue rhythmic gene expression programs differed in breadth, covering global and tissue-specific functions, including metabolic pathways and systemic responses such as heat shock. The circadian clock structure and amplitude was indistinguishable across sexes and age groups. However, overall gene expression rhythms were highly sex-dimorphic and more sustained in females. Moreover, rhythmic programs dampened with age across the body. Together, our stratified analysis unveiled a rich organization of sex- and age-specific circadian gene expression rhythms in humans. One sentence summary Circadian phase inference of 914 GTEx donors reveals sex- and age-dependent rhythmic gene expression programs across 46 human tissues.