Circadian rhythms are fundamental for understanding biology: they date back to the origin of life, they are found in virtually every species from cyanobacteria to mammals, and they coordinate many important biological functions from the sleep-wake cycle, to metabolism, and to cognitive functions. Circadian rhythms are equally fundamental for health and medicine: modifications in diet have been linked to modification in circadian rhythms at the molecular level; disruptions of circadian rhythms have been linked to health problems ranging from depression, to learning disorders, to diabetes, to obesity, to cardiovascular disease, to cancer, and to premature ageing; finally, a large fraction of drug targets have been found to oscillate in a circadian manner in one or several tissues, suggesting that a better understanding of circadian oscillations at the molecular level could have direct applications to precision medicine, for instance by optimizing the time at which drugs are taken. To better understand circadian oscillations at the molecular level, modern high-throughput technologies are being used to measure the concentrations of many molecular species, including transcripts, proteins, and metabolites along the circadian cycle in different organs and tissues, and under different conditions. However, the informatics tools for processing, analyzing, and integrating the growing wealth of molecular circadian data are not yet in place. This effort will fill this fundamental gap by developing and disseminating informatics tools that will enable the collection, integration, and analyses of this wealth of information and lead to novel and fundamental insights about the organization and regulation of circadian oscillations, their roles in health and disease, and their future applications to precision medicine. Specifically, through a close collaborations between computational and experimental scientists, this effort will: (1) Bring the power of deep learning methods to bear on the analyses of omic time series to determine, for instance, which molecular species are oscillating, their characteristics (period, phase, amplitude), and to predict the time/phase associated with a measurement taken at a single time point; (2) Develop Cyber-TC, an extension of the widely used Cyber-T software, for the differential analysis of circadian omic time series and expand MotifMap, a widely used genome-wide map of regulatory sites to better understand circadian regulation; and (3) Develop Circadiomics, an integrated database and web portal as a one-stop shop for circadian data, annotations, and analyses. All data, software, and results will be freely available for academic research purposes and broadly disseminated through multiple channels to benefit both the circadian community and the broader bioinformatics community.
Circadian rhythms are fundamental for biology and medicine. Modern high-throughput technologies are revealing how the concentrations of many molecular species, including transcripts, proteins, and metabolites oscillate with the day and night cycle in almost every species, tissue, and cell. In close collaboration with biologists, this project will develop the informatics tools that will enable the collection, integration, and analyses of this wealth of information and lead to novel and fundamental insights about the organization and regulation of circadian oscillations, their roles in health and disease, and their future applications to precision medicine.
|Urban, Gregor; Tripathi, Priyam; Alkayali, Talal et al. (2018) Deep Learning Localizes and Identifies Polyps in Real Time With 96% Accuracy in Screening Colonoscopy. Gastroenterology 155:1069-1078.e8|
|Baldi, P; Sadowski, P (2018) Learning in the machine: Recirculation is random backpropagation. Neural Netw 108:479-494|
|Kwapis, Janine L; Alaghband, Yasaman; Kramár, Enikö A et al. (2018) Epigenetic regulation of the circadian gene Per1 contributes to age-related changes in hippocampal memory. Nat Commun 9:3323|