MOSAIC: a joint modeling methodology for combined circadian and non-circadian analysis of multi-omics data

Abstract

Motivation Circadian rhythms are approximately 24 hour endogenous cycles that control many biological functions. To identify these rhythms, biological samples are taken over circadian time and analyzed using a single omics type, such as transcriptomics or proteomics. By comparing data from these single omics approaches, it has been shown that transcriptional rhythms are not necessarily conserved at the protein level, implying extensive circadian post-transcriptional regulation. However, as proteomics methods are known to be noisier than transcriptomic methods, this suggests that previously identified arrhythmic proteins with rhythmic transcripts could have been missed due to noise and may not be due to post-transcriptional regulation. Results To determine if one can use information from less-noisy transcriptomic data to inform rhythms in more-noisy proteomic data, and thus more accurately identify rhythms in the proteome, we have created the MOSAIC (Multi-Omics Selection with Amplitude Independent Criteria) application. MOSAIC combines model selection and joint modeling of multiple omics types to recover significant circadian and non-circadian trends. Using both synthetic data and proteomic data from Neurospora crassa, we showed that MOSAIC accurately recovers circadian rhythms at higher rates in not only the proteome but the transcriptome as well, outperforming existing methods for rhythm identification. In addition, by quantifying non-circadian trends in addition to circadian trends in data, our methodology allowed for the recognition of the diversity of circadian regulation as compared to non-circadian regulation. Availability MOSAIC’s full interface is available at https://github.com/delosh653/MOSAIC. An R package for this functionality, mosaic.find, can be downloaded at https://CRAN.R-project.org/package=mosaic.find. Supplementary information Supplementary data are available at Bioinformatics online.

Tick-Tock Consider the Clock: The Influence of Circadian and External Cycles on Time of Day Variation in the Human Metabolome—A Review

“…Many detection algorithms or combination of algorithms have been proposed. Two of these have already been mentioned above (i.e., cosinor and MetaCycle) but many other workflows and applications such as Multi-Omics Selection with Amplitude Independent Criteria (MO-SAIC) [115], Rhythmicity Analysis Incorporating Non-parametric methods (RAIN) [116], Extended Circadian Harmonic Oscillator (ECHO) [117] and others [118,119] are available. As any one method is open to critique, MetaCycle, for example, is already combining a number of different methods to determine different regulation of rhythmicity in different groups of a study or between studies.…”

The PAICE suite reveals circadian posttranscriptional timing of noncoding RNAs and spliceosome components in Mus musculus macrophages

“…The ECHO Linear trend models genes that exhibit both oscillation and a linear trend, with oscillations that increase in a linear manner, possibly due to coregulation of noncircadian and circadian mechanisms. Finally, ECHO Linear Joint models demonstrate the above-described ECHO Linear trend but can only be detected by the joint modeling of the 2 related omic data sets with MOSAIC ( De los Santos et al 2021 ). To use MOSAIC, the LIMBR-adjusted omics data from the more and less noisy data types must be inputted through the MOSAIC “Finding Rhythms” tab.…”

“…1 ). The PAICE suite is designed to identify, visualize, analyze, and contextualize circadian rhythms in a high-throughput fashion in the context of large omics data sets ( De los Santos et al 2019 , 2020 , 2021 ). This suite of R programs, freely available on GitHub ( https://github.com/delosh653/ECHO , https://github.com/delosh653/ENCORE , https://github.com/delosh653/MOSAIC ) and as R packages ( https://cran.r-project.org/web/packages/echo.find/vignettes/echo-vignette.html , https://cran.r-project.org/web/packages/mosaic.find/vignettes/mosaic-vignette.html ), is operated via web-browser-based shiny apps that offer a variety of point-and-click options to assist in the ease of use, allowing users to customize the data analysis options to best suit their data and interests.…”

“…One hypothesis is that the disruption of circadian timing causes the dysregulation of the immune system, which in turn increases inflammation, thereby affecting many physiological processes ( Comas et al 2017 ; Collins et al 2021 ). The most common molecular approach to infer disrupted circadian processes is to identify mRNA, protein, and other key elements of a cell that oscillate over circadian time using omics analysis ( De los Santos et al 2019 , 2020 , 2021 ). Analyzing these omics data to classify which elements in a cell are under the regulation of the circadian clock is a complex and ever-evolving task ( Hughes et al 2010 , 2017 ).…”

“…In response to this issue, we created programs using models that can independently identify oscillations with ACs (Extended Circadian Harmonic Oscillator or ECHO), analyze the ontologies of each AC category (ECHO Native Circadian Ontological Rhythmicity Explorer or ENCORE), and use less noisy data to identify circadian oscillations in corresponding more noisy data (Multi-Omics Selection with Amplitude Independent Criteria or MOSAIC) ( De los Santos et al 2017 , 2019 , 2020 , 2021 ). These individual programs have the capability to more accurately report the classification of genes and cellular pathways that undergo circadian regulation ( De los Santos et al 2020 ; Collins et al 2021 ).…”