See the PhD Opportunities section to see if this project is currently open for applications via MIBTP.
Please Note: The main page lists projects via BBSRC Research Theme(s) quoted and then relevant Topic(s).
Establishing TimeTeller as a tool for human chronotyping
Secondary Supervisor(s): Professor David A Rand (Mathematics)
University of Registration: University of Warwick
BBSRC Research Themes:
Project Outline
Human physiology is governed by the circadian clock, inducing rhythms in key pathways over the 24-day. For example, sleep-wake cycles as well as rhythms in hormones and physiological parameters such as heart rate, blood pressure or blood glucose homeostasis are modulated by the circadian clock. The underlying molecular mechanism is well described. Each cell contains an oscillator driven by transcriptional-(post)translational feedback loops that lead to 20-50% of the transcriptome being expressed with a circadian rhythm. All these cellular clocks within the human organism are governed by the suprachiasmatic nucleus of the hypothalamus, which is considered the central oscillator aligning the physiological rhythms with the environment day/night cycle.
The biological clock is increasingly recognised as a fundamental process important for human health and clock disruption has been shown to negatively impact development of chronic disease but is also a biomarker of ageing as well as some chronic diseases and has already shown some promise as treatment target in some cases [1]. Fully understanding and utilising this “circadian trifactor” is critically dependent on being able to measure the clock in a meaningful and feasible way. Currently, this is done via questionnaires, wearables and the very labour-intensive collection of serial urine or saliva samples for determination of melatonin onset.
Here, we want to build on our longstanding interest in biomoledular markers of the circadian clock [2] and preliminary data suggesting that HairTimeTeller (HTT) is a potential minimally invasive clock biomarker (Figure 1). In fact, hair follicles have long-been suggested as appropriate matrix for clock biomarkers, but given the interindividual variation in gene expression it was not possible to determine the clock from a single sample before.
TimeTeller allows to approach this problem, and an initial model from a time-series of hair follicle RNA is available and a first small-scale validation study in healthy volunteers against a widely accepted phase marker in collaboration with Hattersley (UHCW) is under way. However, further work on a more diverse population at larger scale is needed and will be collected in the course of the PhD. Moreover, an updated version of TimeTeller’s underlying machine learning algorithm is envisioned to advance the method to become self-learning. This multi-disciplinary collaboration has the potential to establish a much-needed novel biomolecular marker for “chronotyping”, potentially enabling population wide studies of circadian clocks.
Figure 1. (A) TimeTeller can detect ‘molecular chronotype’ from transcriptomic analysis of a single oral mucosa sample19 (blue dots individual biopsies, box plots median and interquartile range of each sample for each individual) [see 2]. (B&C) TimeTeller model based on RNA from human hair follicles using nanostring assay accurately predicts phase (chronotype, B) and clock disruption (C) as evident from 1 week of wrist accelerometery in the same subjects. Blue line is the average sleep phase for subject IV which is multiple hours before bedtime in subject I. In C, note the increase in activity during the sleep phase suggesting increased disruption.
References
1. Acosta-Rodríguez, et al. Nat Commun doi: 10.1038/s41467-021-22922-6.
2. Vlachou et al, PLoS Comp Biol doi: 10.1371/journal.pcbi.1011779.
