Different effects of fast and slow input fluctuations on output in gene regulation.

An important task in the post-gene era is to understand the role of stochasticity in gene regulation. Here, we analyze a cascade model of stochastic gene expression, where the upstream gene stochastically generates proteins that regulate, as transcription factors, stochastic synthesis of the downstream output. We find that in contrast to fast input fluctuations that do not change the behavior of the downstream system qualitatively, slow input fluctuations can induce different modes of the distribution of downstream output and even stochastic focusing or defocusing of the downstream output level, although the regulatory protein follows the same distribution in both cases. This finding is counterintuitive but can have broad biological implications, e.g., slow input rather than fast fluctuations may both increase the survival probability of cells and enhance the sensitivity of intracellular regulation. In addition, we find that input fluctuations can minimize the output noise.

Decomposition and tunability of expression noise in the presence of coupled feedbacks.

Expression noise results in cell-to-cell variability in expression levels, and feedback regulation may complicate the tracing of sources of this noise. Using a representative model of gene expression with feedbacks, we analytically show that the expression noise (or the total noise) is decomposed into three parts: feedback-dependent promoter noise determined by a continuous approximation, birth-death noise determined by a simple Poisson process, and correlation noise induced by feedbacks. We clarify confused relationships between feedback and noise in previous studies, by showing that feedback-regulated noisy sources have different contributions to the total noise in different cases of promoter switching (it is an essential reason resulting in confusions). More importantly, we find that there is a tradeoff between response time and expression noise. In addition, we show that in contrast to single feedbacks, coupled positive and negative feedbacks can perform better in tuning expression noise, controlling expression levels, and maintaining response time. The overall analysis implies that living organisms would utilize coupled positive and negative feedbacks for better survival in complex and fluctuating environments.

Fundamental principles of energy consumption for gene expression.

How energy is consumed in gene expression is largely unknown mainly due to complexity of non-equilibrium mechanisms affecting expression levels. Here, by analyzing a representative gene model that considers complexity of gene expression, we show that negative feedback increases energy consumption but positive feedback has an opposite effect; promoter leakage always reduces energy consumption; generating more bursts needs to consume more energy; and the speed of promoter switching is at the cost of energy consumption. We also find that the relationship between energy consumption and expression noise is multi-mode, depending on both the type of feedback and the speed of promoter switching. Altogether, these results constitute fundamental principles of energy consumption for gene expression, which lay a foundation for designing biologically reasonable gene modules. In addition, we discuss possible biological implications of these principles by combining experimental facts.

Channel-Specific and Spatial Residual Attention Network for Medical Image Denoising.

Medical image quality is crucial for physicians to ensure accurate diagnosis and therapeutic strategies. However, due to the interference of noise, there are often various types of noise and artifacts in medical images. This not only damages the visual clarity of images, but also reduces the accuracy of information extraction. Considering that the edges of medical images are rich in high-frequency information, to enhance the quality of medical images, a dual attention mechanism, the channel-specific and spatial residual attention network (CSRAN) in the U-Net framework is proposed. The CSRAN seamlessly integrates the U-Net architecture with channel-wise and spatial feature attention (CSAR) modules, as well as low-frequency channel attention modules. Combined with the two modules, the ability of medical image processing to extract high-frequency features is improved, thereby significantly improving the edge effects and clarity of reconstructed images. This model can present better performance in capturing high-frequency information and spatial structures in medical image denoising and super-resolution reconstruction tasks. It cannot only enhance the ability to extract high-frequency features and strengthen its nonlinear representation capability, but also endow strong edge detection capabilities of the model. The experimental results further prove the superiority of CSRAN in medical image denoising and super-resolution reconstruction tasks.

Architecture-dependent robustness and bistability in a class of genetic circuits.

Understanding the relationship between genotype and phenotype is a challenge in systems biology. An interesting yet related issue is why a particular circuit topology is present in a cell when the same function can supposedly be obtained from an alternative architecture. Here we analyzed two topologically equivalent genetic circuits of coupled positive and negative feedback loops, named NAT and ALT circuits, respectively. The computational search for the oscillation volume of the entire biologically reasonable parameter region through large-scale random samplings shows that the NAT circuit exhibits a distinctly larger fraction of the oscillatory region than the ALT circuit. Such a global robustness difference between two circuits is supplemented by analyzing local robustness, including robustness to parameter perturbations and to molecular noise. In addition, detailed dynamical analysis shows that the molecular noise of both circuits can induce transient switching of the different mechanism between a stable steady state and a stable limit cycle. Our investigation on robustness and dynamics through examples provides insights into the relationship between network architecture and its function.

LogSum + L2 penalized logistic regression model for biomarker selection and cancer classification.

Biomarker selection and cancer classification play an important role in knowledge discovery using genomic data. Successful identification of gene biomarkers and biological pathways can significantly improve the accuracy of diagnosis and help machine learning models have better performance on classification of different types of cancer. In this paper, we proposed a LogSum + L2 penalized logistic regression model, and furthermore used a coordinate decent algorithm to solve it. The results of simulations and real experiments indicate that the proposed method is highly competitive among several state-of-the-art methods. Our proposed model achieves the excellent performance in group feature selection and classification problems.

Physical limits of feedback noise-suppression in biological networks.

Feedback is a ubiquitous control mechanism of biological networks, and has also been identified in a variety of regulatory systems and organisms. It has been shown that, for a given gain and with negligible intrinsic noise, negative feedback impairs noise buffering whereas positive feedback enhances noise buffering. We further investigate the influence of negative and positive feedback on noise in output signals by considering both intrinsic and extrinsic noise as well as operator noise. We find that, while maintaining the system sensitivity, either there exists a minimum of the output noise intensity corresponding to a biologically feasible feedback strength, or the output noise intensity is a monotonic function of feedback strength bounded by both biological and dynamical constraints. In both cases, feedback noise-suppression is physically limited. In other words, noise suppressed by negative or positive feedback cannot be reduced without limitation even in the case of slow transcription.

Geometric characteristics of dynamic correlations for combinatorial regulation in gene expression noise.

Knowing which mode of combinatorial regulation (typically, AND or OR logic operation) that a gene employs is important for determining its function in regulatory networks. Here, we introduce a dynamic cross-correlation function between the output of a gene and its upstream regulator concentrations for signatures of combinatorial regulation in gene expression noise. We find that such a correlation function with respect to the correlation time near the peak close to the point of the zero correlation time is always upward convex in the case of AND logic whereas is always downward convex in the case of OR logic, whichever sources of noise (intrinsic or extrinsic or both). In turn, this fact implies a means for inferring regulatory synergies from available experimental data. The extensions and applications are discussed.

Synchronization and clustering of synthetic genetic networks: a role for cis-regulatory modules.

The effect of signal integration through cis-regulatory modules (CRMs) on synchronization and clustering of populations of two-component genetic oscillators coupled with quorum sensing is investigated in detail. We find that the CRMs play an important role in achieving synchronization and clustering. For this, we investigate six possible cis-regulatory input functions with AND, OR, ANDN, ORN, XOR, and EQU types of responses in two possible kinds of cell-to-cell communications: activator-regulated communication (i.e., the autoinducer regulates the activator) and repressor-regulated communication (i.e., the autoinducer regulates the repressor). Both theoretical analysis and numerical simulation show that different CRMs drive fundamentally different cellular patterns, such as complete synchronization, various cluster-balanced states and several cluster-nonbalanced states.

Coherence, collective rhythm, and phase difference distribution in populations of stochastic genetic oscillators with cellular communication.

An ensemble of stochastic genetic relaxation oscillators via phase-attractive or repulsive cell-to-cell communication are investigated. In the phase-attractive coupling case, it is found that cellular communication can enhance self-induced stochastic resonance as well as collective rhythms, and that different intensities of noise resulting from the fluctuation of intrinsic chemical reactions or the extrinsic environment can induce stochastic limit cycles with different amplitudes for a large cell density. In contrast, in the phase-repulsive coupling case, the distribution of phase differences among the stochastic oscillators can display such characteristic as unimodality, bimodality or polymodality, depending on both noise intensity and cell number, but the modality of phase difference distribution almost keeps invariant for an arbitrary noise intensity as the cell number is beyond a threshold.

Interacting stochastic oscillators.

Stochastic coherence (SC) and self-induced stochastic resonance (SISR) are two distinct mechanisms of noise-induced coherent motion. For interacting SC and SISR oscillators, we find that whether or not phase synchronization is achieved depends sensitively on the coupling strength and noise intensities. Specifically, in the case of weak coupling, individual oscillators are insensitive to each other, whereas in the case of strong coupling, one fixed oscillator with optimal coherence can be entrained to the other, adjustable oscillator (i.e., its noise intensity is tunable), achieving phase-locking synchronization, as long as the tunable noise intensity is not beyond a threshold; such synchronization is lost otherwise. For an array lattice of SISR oscillators, except for coupling-enhanced coherence similar to that found in the case of coupled SC oscillators, there is an optimal network topology degree (i.e., number of coupled nodes), such that coherence and synchronization are optimally achieved, implying that the system-size resonance found in an ensemble of noise-driven bistable systems can occur in coupled SISR oscillators.

Synchronization of genetic oscillators.

Synchronization of genetic or cellular oscillators is a central topic in understanding the rhythmicity of living organisms at both molecular and cellular levels. Here, we show how a collective rhythm across a population of genetic oscillators through synchronization-induced intercellular communication is achieved, and how an ensemble of independent genetic oscillators is synchronized by a common noisy signaling molecule. Our main purpose is to elucidate various synchronization mechanisms from the viewpoint of dynamics, by investigating the effects of various biologically plausible couplings, several kinds of noise, and external stimuli. To have a comprehensive understanding on the synchronization of genetic oscillators, we consider three classes of genetic oscillators: smooth oscillators (exhibiting sine-like oscillations), relaxation oscillators (displaying jump dynamics), and stochastic oscillators (noise-induced oscillation). For every class, we further study two cases: with intercellular communication (including phase-attractive and repulsive coupling) and without communication between cells. We find that an ensemble of smooth oscillators has different synchronization phenomena from those in the case of relaxation oscillators, where noise plays a different but key role in synchronization. To show differences in synchronization between them, we make comparisons in many aspects. We also show that a population of genetic stochastic oscillators have their own synchronization mechanisms. In addition, we present interesting phenomena, e.g., for relaxation-type stochastic oscillators coupled to a quorum-sensing mechanism, different noise intensities can induce different periodic motions (i.e., inhomogeneous limit cycles).

The free-energy cost of interaction between DNA loops.

From the viewpoint of thermodynamics, the formation of DNA loops and the interaction between them, which are all non-equilibrium processes, result in the change of free energy, affecting gene expression and further cell-to-cell variability as observed experimentally. However, how these processes dissipate free energy remains largely unclear. Here, by analyzing a mechanic model that maps three fundamental topologies of two interacting DNA loops into a 4-state model of gene transcription, we first show that a longer DNA loop needs more mean free energy consumption. Then, independent of the type of interacting two DNA loops (nested, side-by-side or alternating), the promotion between them always consumes less mean free energy whereas the suppression dissipates more mean free energy. More interestingly, we find that in contrast to the mechanism of direct looping between promoter and enhancer, the facilitated-tracking mechanism dissipates less mean free energy but enhances the mean mRNA expression, justifying the facilitated-tracking hypothesis, a long-standing debate in biology. Based on minimal energy principle, we thus speculate that organisms would utilize the mechanisms of loop-loop promotion and facilitated tracking to survive in complex environments. Our studies provide insights into the understanding of gene expression regulation mechanism from the view of energy consumption.

Mechanisms of information decoding in a cascade system of gene expression.

Biotechnology advances have allowed investigation of heterogeneity of cellular responses to stimuli on the single-cell level. Functionally, this heterogeneity can compromise cellular responses to environmental signals, and it can also enlarge the repertoire of possible cellular responses and hence increase the adaptive nature of cellular behaviors. However, the mechanism of how this response heterogeneity is generated remains elusive. Here, by systematically analyzing a representative cellular signaling system, we show that (1) the upstream activator always amplifies the downstream burst frequency (BF) but the noiseless activator performs better than the noisy one, remarkably for small or moderate input signal strengths, and the repressor always reduces the downstream BF but the difference in the reducing effect between noiseless and noise repressors is very small; (2) both the downstream burst size and mRNA mean are a monotonically increasing function of the activator strength but a monotonically decreasing function of the repressor strength; (3) for repressor-type input, there is a noisy signal strength such that the downstream mRNA noise arrives at an optimal level, but for activator-type input, the output noise intensity is fundamentally a monotonically decreasing function of the input strength. Our results reveal the essential mechanisms of both signal information decoding and cellular response heterogeneity, whereas our analysis provides a paradigm for analyzing dynamics of noisy biochemical signaling systems.

Correction: Division time-based amplifiers for stochastic gene expression.

Correction for 'Division time-based amplifiers for stochastic gene expression' by Haohua Wang et al., Mol. BioSyst., 2015, 11, 2417-2428.

Effects of promoter leakage on dynamics of gene expression.

BACKGROUND: Quantitative analysis of simple molecular networks is an important step forward understanding fundamental intracellular processes. As network motifs occurring recurrently in complex biological networks, gene auto-regulatory circuits have been extensively studied but gene expression dynamics remain to be fully understood, e.g., how promoter leakage affects expression noise is unclear. RESULTS: In this work, we analyze a gene model with auto regulation, where the promoter is assumed to have one active state with highly efficient transcription and one inactive state with very lowly efficient transcription (termed as promoter leakage). We first derive the analytical distribution of gene product, and then analyze effects of promoter leakage on expression dynamics including bursting kinetics. Interestingly, we find that promoter leakage always reduces expression noise and that increasing the leakage rate tends to simplify phenotypes. In addition, higher leakage results in fewer bursts. CONCLUSIONS: Our results reveal the essential role of promoter leakage in controlling expression dynamics and further phenotype. Specifically, promoter leakage is a universal mechanism of reducing expression noise, controlling phenotypes in different environments and making the gene produce generate fewer bursts.

Roles of factorial noise in inducing bimodal gene expression.

Some gene regulatory systems can exhibit bimodal distributions of mRNA or protein although the deterministic counterparts are monostable. This noise-induced bimodality is an interesting phenomenon and has important biological implications, but it is unclear how different sources of expression noise (each source creates so-called factorial noise that is defined as a component of the total noise) contribute separately to this stochastic bimodality. Here we consider a minimal model of gene regulation, which is monostable in the deterministic case. Although simple, this system contains factorial noise of two main kinds: promoter noise due to switching between gene states and transcriptional (or translational) noise due to synthesis and degradation of mRNA (or protein). To better trace the roles of factorial noise in inducing bimodality, we also analyze two limit models, continuous and adiabatic approximations, apart from the exact model. We show that in the case of slow gene switching, the continuous model where only promoter noise is considered can exhibit bimodality; in the case of fast switching, the adiabatic model where only transcriptional or translational noise is considered can also exhibit bimodality but the exact model cannot; and in other cases, both promoter noise and transcriptional or translational noise can cooperatively induce bimodality. Since slow gene switching and large protein copy numbers are characteristics of eukaryotic cells, whereas fast gene switching and small protein copy numbers are characteristics of prokaryotic cells, we infer that eukaryotic stochastic bimodality is induced mainly by promoter noise, whereas prokaryotic stochastic bimodality is induced primarily by transcriptional or translational noise.

Division time-based amplifiers for stochastic gene expression.

While cell-to-cell variability is a phenotypic consequence of gene expression noise, sources of this noise may be complex - apart from intrinsic sources such as the random birth/death of mRNA and stochastic switching between promoter states, there are also extrinsic sources of noise such as cell division where division times are either constant or random. However, how this time-based division affects gene expression as well as how it contributes to cell-to-cell variability remains unexplored. Using a computational model combined with experimental data, we show that the cell-cycle length defined as the difference between two sequential division times can significantly impact the expression dynamics. Specifically, we find that both divisions (constant or random) always increase the mean level of mRNA and lengthen the mean first passage time. In contrast to constant division, random division always amplifies expression noise but tends to stabilize its temporal level, and unimodalizes the mRNA distribution, but makes its tail longer. These qualitative results reveal that cell division based on time is an effective mechanism for both increasing expression levels and enhancing cell-to-cell variability.

Reinitiation enhances reliable transcriptional responses in eukaryotes.

Gene transcription is a noisy process carried out by the transcription machinery recruited to the promoter. Noise reduction is a fundamental requirement for reliable transcriptional responses which in turn are crucial for signal transduction. Compared with the relatively simple transcription initiation in prokaryotes, eukaryotic transcription is more complex partially owing to its additional reinitiation mechanism. By theoretical analysis, we showed that reinitiation reduces noise in eukaryotic transcription independent of the transcription level. Besides, a higher reinitiation rate enables a stable scaffold complex an advantage in noise reduction. Finally, we showed that the coupling between scaffold formation and transcription can further reduce transcription noise independent of the transcription level. Furthermore, compared with the reinitiation mechanism, the noise reduction effect of the coupling can be of more significance in the case that the transcription level is low and the intrinsic noise dominates. Our results uncover a mechanistic route which eukaryotes may use to facilitate a more reliable response in the noisy transcription process.

Feedback-induced counterintuitive correlations of gene expression noise with bursting kinetics.

Previous studies showed that a higher frequency of bursting results in lower expression noise whereas a larger size of bursting leads to higher expression noise. Here, we show counterintuitive correlations of expression noise with bursting kinetics due to the effect of feedback. Specifically, in the case of increasing the negative feedback strength but keeping the mean expression fixed, both the mean burst frequency and the mean burst size are invariant if the off-switching rate decreases, but expression noise is reduced; or the mean burst frequency is invariant and the burst size decreases if the transcription rate increases, but expression noise is amplified. Similarly, in the case of increasing the positive feedback strength but keeping the mean expression fixed, both the mean burst frequency and the mean burst size are invariant if the on-switching rate decreases; or the mean burst frequency increases and the mean burst size is invariant if the leakage rate decreases, but expression noise is amplified. In addition, we find that the previous conclusion that a larger burst size results in the lower noise in burst size needs to be modified in the case of feedback. Our results not only clarify the confusing relationship between feedback and expression noise but also imply that the mRNA or protein noise is no longer a simple sum of the internal noise and the promoter noise as shown in the case of no feedback.