17 Metabolic oscillations percolate throughout cellular physiology. The transcriptome oscillates 18 between expression of genes encoding for biosynthesis and growth, and for catabolism and stress 19 response. Long protein half-lives suppress effects on protein abundance level, and the function of 20 periodic transcription remains elusive. We performed RNA-seq analysis during a dynamic state of 21 the system. Short protein half-lives and high transcript abundance amplitude of sulfur uptake 22 genes and carbonic anhydrase, and dynamic changes of pathway intermediates H2S and CO2 23 support a direct role of transcription in cycle dynamics. Substantial changes in the relative duration 24 of expression of the antagonistic co-expression cohorts precede a system bifurcation to a longer 25 period, supporting the idea of a function in cellular resource allocation. The pulse-width 26 modulation model, a mathematical formulation of this idea, can explain a large body of published 27 experimental data on the dependence of the cycle period on the culture growth rate. This 28 pulse-like model of cell growth provides a first theoretical framework, where the phenomenon is 29 understood as a mechanism of cellular resource allocation and protein homeostasis, and is 30 applicable to circadian transcriptome dynamics from all domains of life. 31

16 Metabolic oscillation percolate throughout cellular physiology. The transcriptome oscillates 17 between gene cohorts encoding for biosynthesis and growth, and for catabolism and stress 18 response. However, long protein half-lives dampen effects on protein abundance, and the 19 function of periodic transcription remains elusive. To test three prevailing hypotheses, we 20 performed RNA-seq analysis during a dynamic state of the system, just preceding a bifurcation of 21 system dynamics to a longer period. A few enzymes with short protein half-live and high 22 transcript abundance amplitude could play a direct role in metabolic dynamics (hypothesis H1), 23 e.g., carbonic anhydrase, sulfate uptake and the glyoxylate cycle. Substantial changes in the 24 relative duration of expression of the antagonistic cohorts precede the bifurcation, supporting 25 the idea of a function in cellular resource allocation (H2). The pulse-width modulation model, a 26 mathematical formulation of this idea, explains the dependence of the cycle period on the culture 27 growth rate. This pulse-like model of cell growth provides a first theoretical framework, where 28 the phenomenon is understood as a mechanism of cellular resource allocation and protein 29 homeostasis. The universal temporal program of transcription encodes the spatial structure of 30 the cell, and we further suggest a function in pattern formation of the growing cell (H3). 31

The problem of segmenting linearly ordered data is frequently encountered in time-series analysis, computational biology, and natural language processing. Segmentations obtained independently from replicate data sets or from the same data with different methods or parameter settings pose the problem of computing an aggregate or consensus segmentation. This Segmentation Aggregation problem amounts to finding a segmentation that minimizes the sum of distances to the input segmentations. It is again a segmentation problem and can be solved by dynamic programming. The aim of this contribution is (1) to gain a better mathematical understanding of the Segmentation Aggregation problem and its solutions and (2) to demonstrate that consensus segmentations have useful applications. Extending previously known results we show that for a large class of distance functions only breakpoints present in at least one input segmentation appear in the consensus segmentation. Furthermore, we derive a bound on the size of consensus segments. As show-case applications, we investigate a yeast transcriptome and show that consensus segments provide a robust means of identifying transcriptomic units. This approach is particularly suited for dense transcriptomes with polycistronic transcripts, operons, or a lack of separation between transcripts. As a second application, we demonstrate that consensus segmentations can be used to robustly identify growth regimes from sets of replicate growth curves.

Metabolic oscillations are characterized by alternating phases of high and low respiratory activity, associated with transcription of genes involved in biosynthetic pathways and growth, and in catabolism and stress response. However, the functional consequences of transcriptome oscillations remain unclear, since most proteins are too stable to be affected by oscillatory transcript abundances. In this work, we investigate a transcriptome time series during an unstable state of the oscillation. Our analyses confirm previous suggestions that the relative times spent in the alternative transcription states are coupled to growth rate. This pulse-width modulation of transcription provides a simple mechanism for the long-standing question of how cells adjust their ribosome content and growth rate to environmental conditions. A mathematical model of this idea reproduces both the almost linear relation of transcript and protein abundances and the non-linear relation of oscillation periods to growth rate.

In the 1950s, parallel to revolutions in our understanding of DNA and the cell division cycle (CDC), there were also huge developments in biotechnology. As some of the first domesticated microbes, Saccharomyces spp. have been at the vanguard of these developments. Specifically, the development of continuous culture has shaped cell growth laws and our understanding of metabolism. As early as 1954, Finn and his coworkers observed that during the continuous growth of Saccharomyces pastorianus (lager yeast) an autonomous oscillation in fermentation products, pH and dissolved oxygen developed during steady-state conditions, providing a first taste of metabolic and CDC integration.1 As technology advanced, Finn’s observations were rediscovered and added to, using the stalwart of biotechnology bakers’ yeast (Saccharomyces cerevisiae). Von Meyenburg refined culture conditions and precisely measured growth kinetics, gas exchanges and energetics to define the relationship between growth, the CDC and respiration (budding commences at high respiration rates).2 The Fiechter group closely tracked DNA synthesis using flow cytometry to show that S-phase occurred during high respiration.3 The periods observed during the oscillation were about half the doubling time of the culture (4∼6h) so the oscillation was thought to be an inherent property of the CDC and respiratory capacity. In 1991, the Kuriyama group showed that a much shorterperiod respiratory oscillation (usually ranging between 40∼60 min) could develop under similar growth conditions.4 In the following years, there were many insights from the groups of Kuriyama, Lloyd and Murray. 5 Further studies used growth with ethanol, measurement of glutathione levels and by inhibitors of its synthesis, and inhibition by nitrosation reagents. Continuous monitoring of NAD(P)H and flavin fluorescence showed that the oscillation could be entirely respiratory and did not require either glucose, fermentation or glycogen accumulation. Therefore by 2005, the oscillation phases could be defined according to flux through the mitochondrial electron transport chain (ETC), showing respiratory control (ADP-acceptor control), and also responsive to uncouplers of mitochondrial energy conservation. The respiratory/oxidative (high oxygen consumption; HOC) phase has a lower residual dissolved oxygen concentration (Figure 1A) than the reductive phase (low oxygen consumption; LOC). Residual dissolved oxygen concentration exactly tracks the oxygen uptake rate (and thus ETC flux) because continuous culture is an open system (Figure 1 B), i.e., nutrients such as oxygen, are continuously perfused into the reactor, so any changes in nutrient concentration in the reactor are indicative of flux changes. Population synchrony was shown to be mediated by small molecules such as acetaldehyde, as shown by monitoring phase response curves. The period of the oscillation is temperature compensated, and period lengthening occurs with Li+, and Atype monoamine oxidase inhibitors (Phenalzine and Iproniazid). These psychotropic drugs are also well known to prolong period circadian timing, and are effectors of phospho-inositide signaling; this suggests a shared pathway between longer and shorter time domains. Most significantly, it became evident that the metabolic oscillation, was not solely a downstream function of the CDC, but a rhythm (an ultradian clock). Up until this point, the oscillation had largely been published in specialist biotechnology and microbiology journals. However,

16 Metabolic oscillations in budding yeast percolate throughout cellular physiology. The transcriptome 17 oscillates between expression of genes encoding for biosynthesis and growth, and for catabolism 18 and stress response. The two prevalent hypotheses on functions of periodic transcript abundances 19 are H1, where transcript abundance oscillation is in direct feedback with metabolism; and H2, 20 where the length of the pulses can adjust relative protein abundances, e.g., of ribosomal proteins, 21 to the growth rate. To test these two hypotheses, we performed RNA-seq analysis of a time series 22 over 2.5 cycles of a short period respiratory oscillation (≈ 36min, strain IFO 0233), followed by a 23 distinct bifurcation of system dynamics. H1 is supported by short protein half-lives and high 24 transcript abundance amplitude of sulfur uptake genes and carbonic anhydrase, as well as by 25 dynamic changes of pathway intermediates H2S and CO2. Substantial changes in the relative 26 duration of expression of the antagonistic co-expression cohorts precede the bifurcation to a 27 longer period, in support of H2. We derive a simple mathematical formulation of H2, the 28 pulse-width modulation model. This model can explain a large body of published experimental 29 data on the dependance of the cycle period on the culture growth rate, and provides a novel 30 theoretical framework where the phenomenon is understood as a resource allocation mechanism. 31 Potential roles in protein homeostasis and subcellular pattern formation (H3) are discussed. In 32 summary, our work consolidates the picture of a eukaryotic cell growth cycle, where pulses of 33 protein biosynthesis shape the growing cell. 34

Britton Chance, electronics expert when a teenager, became an enthusiastic student of biological oscillations, passing on this enthusiasm to many students and colleagues, including one of us (DL). This historical essay traces BC's influence through the accumulated work of DL to DL's many collaborators. The overall temporal organization of mass-energy, information, and signaling networks in yeast in self-synchronized continuous cultures represents, until now, the most characterized example of in vivo elucidation of time structure. Continuous online monitoring of dissolved gases by direct measurement (membrane-inlet mass spectrometry, together with NAD(P)H and flavin fluorescence) gives strain-specific dynamic information from timescales of minutes to hours as does two-photon imaging. The predominantly oscillatory behavior of network components becomes evident, with spontaneously synchronized cellular respiration cycles between discrete periods of increased oxygen consumption (oxidative phase) and decreased oxygen consumption (reductive phase). This temperature-compensated ultradian clock provides coordination, linking temporally partitioned functions by direct feedback loops between the energetic and redox state of the cell and its growing ultrastructure. Multioscillatory outputs in dissolved gases with 13 h, 40 min, and 4 min periods gave statistical self-similarity in power spectral and relative dispersional analyses: i.e., complex nonlinear (chaotic) behavior and a functional scale-free (fractal) network operating simultaneously over several timescales. (C) The Authors. Published by SPIE under a Creative Commons Attribution 3.0 Unported License.

The segmentation of time series and genomic data is a common problem in computational biology. With increasingly complex measurement procedures individual data points are often not just numbers or simple vectors in which all components are of the same kind. Analysis methods that capitalize on slopes in a single real-valued data track or that make explicit use of the vectorial nature of the data are not applicable in such scenaria. We develop here a framework for segmentation in arbitrary data domains that only requires a minimal notion of similarity. Using unsupervised clustering of (a sample of) the input yields an approximate segmentation algorithm that is efficient enough for genome-wide applications. As a showcase application we segment a time-series of transcriptome sequencing data from budding yeast, in high temporal resolution over ca. 2.5 cycles of the short-period respiratory oscillation. The algorithm is used with a similarity measure focussing on periodic expression profiles across the metabolic cycle rather than coverage per time point.

Oscillations play a significant role in biological systems, with many examples in the fast, ultradian, circadian, circalunar and yearly time domains. However, determining periodicity in such data can be problematic. There are a number of computational methods to identify the periodic components in large datasets, such as signal-to-noise based Fourier decomposition, Fisher's g-test and autocorrelation. However, the available methods assume a sinusoidal model and do not attempt to quantify the waveform shape and the presence of multiple periodicities, which provide vital clues in determining the underlying dynamics. Here, we developed a Fourier based measure that generates a de-noised waveform from multiple significant frequencies. This waveform is then correlated with the raw data from the respiratory oscillation found in yeast, to provide oscillation statistics including waveform metrics and multi-periods. The method is compared and contrasted to commonly used statistics. Moreover we show the utility of the program in the analysis of noisy datasets and other high-throughput analyses, such as metabolomics and flow cytometry, respectively.

A plethora of data is accumulating from high throughput methods on metabolites, coenzymes, proteins, lipids and nucleic acids and their interactions as well as the signalling and regulatory functions and pathways of the cellular network. The frozen moment viewed in a single discrete time sample requires frequent repetition and updating before any appreciation of the dynamics of component interaction becomes possible. Even then in a sample derived from a cell population, time-averaging of processes and events that occur in out-of-phase individuals blur the detailed complexity of single cell.Continuously grown cultures of yeast spontaneously self-synchronise and provide resolution of detailed temporal structure. Continuous online monitoring (O-2 electrode and membrane-inlet mass spectrometry for O-2, CO2 and H2S; direct fluorimetry for NAD(P) H and flavins) gives dynamic information from timescales of minutes to hours. When these data are supplemented with mass spectrometry-based metabolomics and transcriptomics, the predominantly oscillatory behaviour of network components becomes evident, where respiration cycles between increased oxygen consumption (oxidative phase) and decreased oxygen consumption (reductive phase). This ultradian clock provides a coordinating function that links mitochondrial energetics and redox balance to transcriptional regulation, mitochondrial structure and organelle remodelling, DNA duplication and chromatin dynamics. Ultimately, anabolism and catabolism become globally partitioned: mediation is by direct feedback loops between the energetic and redox state of the cell and chromatin architecture via enzymatic co-factors and co-enzymes.Multi-oscillatory outputs were observed in dissolved gases with 12-h, 40-min and 4-min periods, and statistical self-similarity in Power Spectral and Relative Dispersional analyses: i.e. complex non-linear behaviour and a functional scale-free network operating simultaneously on several timescales. Fast sampling (at 10 or 1 Hz) of NAD(P) H fluorescence revealed subharmonic components of the 40-min signal at 20,10 and 3-5 min. The latter corresponds to oscillations directly observed and imaged by 2-photon microscopy in surface-attached cells. Signalling between time domains is suggested by studies with protonophore effectors of mitochondrial energetics. Multi-oscillatory states impinge on the complex reactome (where concentrations of most chemical species oscillate) and network functionality is made more comprehensible when in vivo time structure is taken into account.

The structural dynamics of chromatin have been implicated in the regulation of fundamental eukaryotic processes, such as DNA transcription, replication and repair. Although previous studies have revealed that the chromatin landscape, nucleosome remodeling and histone modification events are intimately tied into cellular energetics and redox state, few studies undertake defined time-resolved measurements of these state variables. Here, we use metabolically synchronous, continuously-grown yeast cultures to measure DNA occupancy and track global patterns with respect to the metabolic state of the culture. Combined with transcriptome analyses and ChIP-qPCR experiments, these paint an intriguing picture where genome-wide nucleosome focusing occurs during the recovery of energy charge, followed by clearance of the promoter regions and global transcriptional slow-down, thus indicating a nucleosome-mediated “reset point” for the cycle. The reset begins at the end of the catabolic and stress-response transcriptional programs and ends prior to the start of the anabolic and cell-growth transcriptional program, and the histones on genes from both the catabolic and anabolic superclusters are deacetylated.

A plethora of data is accumulating from high throughput methods on metabolites, coenzymes, proteins, and nucleic acids and their interactions as well as the signalling and regulatory functions and pathways of the cellular network. The frozen moment viewed in a single discrete time sample requires frequent repetition and updating before any appreciation of the dynamics of component interaction becomes possible. Even then in a sample derived from a cell population, time-averaging of processes and events that occur in out-of-phase individuals blur the detailed complexity of single cell organization. Continuously-grown cultures of yeast can become spontaneously self-synchronized, thereby enabling resolution of far more detailed temporal structure. Continuous on-line monitoring by rapidly responding sensors (O-2 electrode and membrane-inlet mass spectrometry for O-2, CO2 and H2S; direct fluorimetry for NAD(P) H and flavins) gives dynamic information from time-scales of minutes to hours. Supplemented with capillary electophoresis and gas chromatography mass spectrometry and transcriptomics the predominantly oscillatory behaviour of network components becomes evident, with a 40 min cycle between a phase of increased respiration (oxidative phase) and decreased respiration (reductive phase). Highly pervasive, this ultradian clock provides a coordinating function that links mitochondrial energetics and redox balance to transcriptional regulation, mitochondrial structure and organelle remodelling, DNA duplication and cell division events. Ultimately, this leads to a global partitioning of anabolism and catabolism and the enzymes involved, mediated by a relatively simple ATP feedback loop on chromatin architecture.

Gas–liquid mass transfer is often rate‐limiting in laboratory and industrial cultures of aerobic or autotrophic organisms. The volumetric mass transfer coefficient kLa is a crucial characteristic for comparing, optimizing, and upscaling mass transfer efficiency of bioreactors. Reliable dynamic models and resulting methods for parameter identification are needed for quantitative modeling of microbial growth dynamics. We describe a laboratory‐scale stirred tank reactor (STR) with a highly efficient aeration system (kLa ≈ 570 h−1). The reactor can sustain yeast culture with high cell density and high oxygen uptake rate, leading to a significant drop in gas concentration from inflow to outflow (by 21%). Standard models fail to predict the observed mass transfer dynamics and to identify kLa correctly. In order to capture the concentration gradient in the gas phase, we refine a standard ordinary differential equation (ODE) model and obtain a system of partial integro‐differential equations (PIDE), for which we derive an approximate analytical solution. Specific reactor configurations, in particular a relatively short bubble residence time, allow a quasi steady‐state approximation of the PIDE system by a simpler ODE model which still accounts for the concentration gradient. Moreover, we perform an appropriate scaling of all variables and parameters. In particular, we introduce the dimensionless overall efficiency κ, which is more informative than kLa since it combines the effects of gas inflow, exchange, and solution. Current standard models of mass transfer in laboratory‐scale aerated STRs neglect the gradient in the gas concentration, which arises from highly efficient bubbling systems and high cellular exchange rates. The resulting error in the identification of κ (and hence kLa) increases dramatically with increasing mass transfer efficiency. Notably, the error differs between cell‐free and culture‐based methods of parameter identification, potentially confounding the determination of the “biological enhancement” of mass transfer. Our new model provides an improved theoretical framework that can be readily applied to aerated bioreactors in research and biotechnology. Biotechnol. Bioeng. 2012; 109: 2997–3006. © 2012 Wiley Periodicals, Inc.

Conventional extraction protocols for yeast have been developed for relatively rapid‐growing low cell density cultures of laboratory strains and often do not have the integrity for frequent sampling of cultures. Therefore, these protocols are usually inefficient for cultures under slow growth conditions or of non‐laboratory strains. We have developed a combined mechanical and chemical disruption procedure using vigorous bead‐beating that can consistently disrupt yeast cells (> 95%), irrespective of cell cycle and metabolic state. Using this disruption technique coupled with quenching, we have developed DNA, RNA and protein extraction protocols that are optimized for a large number of samples from slow‐growing high‐density industrial yeast cultures. Additionally, sample volume, the use of expensive reagents/enzymes, handling times and incubations were minimized. We have tested the reproducibility of our methods using triplicate/time‐series extractions and compared these with commonly used protocols or commercially available kits. Moreover, we utilized a simple flow‐cytometric approach to estimate the mitochondrial DNA copy number. Based on the results, our methods have shown higher reproducibility, yield and quality. Copyright © 2012 John Wiley & Sons, Ltd.

All previous studies on the yeast metabolome have yielded a plethora of information on the components, function and organisation of low molecular mass and macromolecular components involved in the cellular metabolic network. Here we emphasise that an understanding of the global dynamics of the metabolome in vivo requires elucidation of the temporal dynamics of metabolic processes on many time-scales. We illustrate this using the 40 min oscillation in respiratory activity displayed in auto-synchronous continuously grown cultures of Saccharomyces cerevisiae, where respiration cycles between a phase of increased respiration (oxidative phase) and decreased respiration (reductive phase). Thereby an ultradian clock, i.e. a timekeeping device that runs through many cycles during one day, is involved in the co-ordination of the vast majority of events and processes in yeast. Through continuous online measurements, we first show that mitochondrial and redox physiology are intertwined to produce the temporal landscape on which cellular events occur. Next we look at the higher order processes of DNA duplication and mitochondrial structure to reveal that both events are choreographed during the respiratory cycles. Furthermore, spectral analysis using the discrete Fourier transformation of high-resolution (10 Hz) time-series of NAD(P)H confirms the existence of higher frequency components of biological origin and that these follow a scale-free architecture even in stable oscillating modes. A different signal-processing approach using discrete wavelet transformations (DWT) indicates that there is a significant contribution to the overall signal from ∼ 5, ∼ 10 and ∼ 20-minutes cycles and the amplitudes of these cycles are phase-dependent. Further investigation (derivative of Gaussian continuous wavelet transformation) reveals that the observed 20-minutes cycles are actually confined to the reductive phase and consist of two ∼ 15-minutes cycles. Moreover, the 5 and 10-minutes cycles are restricted to the oxidative phase of the cycle. The mitochondrial origin of these signals was confirmed by pulse-injection of the cytochrome c oxidase inhibitor H2S. We next discuss how these multi-oscillatory states can impinge on the apparently complex reactome (represented as a phase diagram of 1,650 chemical species that show oscillatory behaviour). We conclude that biological processes can be considerably more comprehensible when dynamic in vivo time-structure is taken into account.

Bacteria dynamically exchange with their environment by constantly uptaking nutrients and secreting metabolic products and other biomolecules. While such secreted metabolites may represent a high-level reporter of metabolic activity of the culture, relatively few studies have focused on their characterization. In addition, metabolites may be potential mediators of intercellular interactions. This study aims at identifying candidate mediators of intercellular exchanges and population behavior from temporal patterns of metabolites. To do this, we used capillary electrophoresis mass spectrometry (CE-MS) to monitor secreted metabolites in synchronized continuous culture of E. coli displaying respiratory oscillations. We observed that multiple metabolites are secreted in significant quantities in the extracellular medium, including amino acids and other intermediates of central metabolism. Some of the secreted metabolite dynamics appear linked to the known valine toxicity in E. coli and are also associated with the respiratory oscillations and their dynamics. Moreover, the dynamics in the level of several amino acids appeared well correlated, suggesting organized cycles of secretion/reuptake during respiratory and metabolic shifts linked to valine levels. Overall, the current results suggest that multiple metabolites are produced and likely exchanged by E. coli during continuous growth. These appear to reflect the internal metabolic state of the cell and may form an underappreciated level of information exchange that cell populations use to coordinate activities.

When grown in continuous culture, budding yeast cells tend to synchronize their respiratory activity to form a stable oscillation that percolates throughout cellular physiology and involves the majority of the protein-coding transcriptome. Oscillations in batch culture and at single cell level support the idea that these dynamics constitute a general growth principle. The precise molecular mechanisms and biological functions of the oscillation remain elusive. Fourier analysis of transcriptome time series datasets from two different oscillation periods (0.7 h and 5 h) reveals seven distinct co-expression clusters common to both systems (34% of all yeast ORF), which consolidate into two superclusters when correlated with a compilation of 1,327 unrelated transcriptome datasets. These superclusters encode for cell growth and anabolism during the phase of high, and mitochondrial growth, catabolism and stress response during the phase of low oxygen uptake. The promoters of each cluster are characterized by different nucleotide contents, promoter nucleosome configurations, and dependence on ATP-dependent nucleosome remodeling complexes. We show that the ATP:ADP ratio oscillates, compatible with alternating metabolic activity of the two superclusters and differential feedback on their transcription via activating (RSC) and repressive (Isw2) types of promoter structure remodeling. We propose a novel feedback mechanism, where the energetic state of the cell, reflected in the ATP:ADP ratio, gates the transcription of large, but functionally coherent groups of genes via differential effects of ATP-dependent nucleosome remodeling machineries. Besides providing a mechanistic hypothesis for the delayed negative feedback that results in the oscillatory phenotype, this mechanism may underpin the continuous adaptation of growth to environmental conditions.

There is an increasing call for the absolute quantification of time-resolved metabolite data. However, a number of technical issues exist, such as metabolites being modified/degraded either chemically or enzymatically during the extraction process. Additionally, capillary electrophoresis mass spectrometry (CE-MS) is incompatible with high salt concentrations often used in extraction protocols. In microbial systems, metabolite yield is influenced by the extraction protocol used and the cell disruption rate. Here we present a method that rapidly quenches metabolism using dry-ice ethanol bath and methanol N-ethylmaleimide solution (thus stabilising thiols), disrupts cells efficiently using bead-beating and avoids artefacts created by live-cell pelleting. Rapid sample processing minimised metabolite leaching. Cell weight, number and size distribution was used to calculate metabolites to an attomol/cell level. We apply this method to samples obtained from the respiratory oscillation that occurs when yeast are grown continuously.

BACKGROUND:In biological systems, redox reactions are central to most cellular processes and the redox potential of the intracellular compartment dictates whether a particular reaction can or cannot occur. Indeed the widespread use of redox reactions in biological systems makes their detailed description outside the scope of one review. SCOPE OF THE REVIEW:Here we will focus on how system-wide redox changes can alter the reaction and transcriptional landscape of Saccharomyces cerevisiae. To understand this we explore the major determinants of cellular redox potential, how these are sensed by the cell and the dynamic responses elicited. MAJOR CONCLUSIONS:Redox regulation is a large and complex system that has the potential to rapidly and globally alter both the reaction and transcription landscapes. Although we have a basic understanding of many of the sub-systems and a partial understanding of the transcriptional control, we are far from understanding how these systems integrate to produce coherent responses. We argue that this non-linear system self-organises, and that the output in many cases is temperature-compensated oscillations that may temporally partition incompatible reactions in vivo. GENERAL SIGNIFICANCE:Redox biochemistry impinges on most of cellular processes and has been shown to underpin ageing and many human diseases. Integrating the complexity of redox signalling and regulation is perhaps one of the most challenging areas of biology. This article is part of a Special Issue entitled Systems Biology of Microorganisms.