NPJ Systems Biology and Applications最新文献

Mechanistic mathematical models (MMs) are a powerful tool to help us understand and predict the dynamics of tumour growth under various conditions. In this work, we use 5 MMs with an increasing number of parameters to explore how certain (often overlooked) decisions in estimating parameters from data of experimental tumour growth affect the outcome of the analysis. In particular, we propose a framework for including tumour volume measurements that fall outside the upper and lower limits of detection, which are normally discarded. We demonstrate how excluding censored data results in an overestimation of the initial tumour volume and the MM-predicted tumour volumes prior to the first measurements, and an underestimation of the carrying capacity and the MM-predicted tumour volumes beyond the latest measurable time points. We show in which way the choice of prior for the MM parameters can impact the posterior distributions, and illustrate that reporting the most likely parameters and their 95% credible interval can lead to confusing or misleading interpretations. We hope this work will encourage others to carefully consider choices made in parameter estimation and to adopt the approaches we put forward herein.

We present a study where predictive mechanistic modeling is combined with deep learning methods to predict individual patient survival probabilities under immune checkpoint inhibitor (ICI) immunotherapy. This hybrid approach enables prediction based on both measures that are calculable from mechanistic models of key mechanisms underlying ICI therapy that may not be directly measurable in the clinic and easily measurable quantities or patient characteristics that are not always readily incorporated into predictive mechanistic models. A deep learning time-to-event predictive model trained on a hybrid mechanistic + clinical data set from 93 patients achieved higher per-patient predictive accuracy based on event-time concordance, Brier score, and negative binomial log-likelihood-based criteria than when trained on only mechanistic model-derived values or only clinical data. Feature importance analysis revealed that both clinical and model-derived parameters play prominent roles in increasing prediction accuracy, further supporting the advantage of our hybrid approach.

Network controllability is unifying the traditional control theory with the structural network information rooted in many large-scale biological systems of interest, from intracellular networks in molecular biology to brain neuronal networks. In controllability approaches, the set of minimum driver nodes is not unique, and critical nodes are the most important control elements because they appear in all possible solution sets. On the other hand, a common but largely unexplored feature in network control approaches is the probabilistic failure of edges or the uncertainty in the determination of interactions between molecules. This is particularly true when directed probabilistic interactions are considered. Until now, no efficient algorithm existed to determine critical nodes in probabilistic directed networks. Here we present a probabilistic control model based on a minimum dominating set framework that integrates the probabilistic nature of directed edges between molecules and determines the critical control nodes that drive the entire network functionality. The proposed algorithm, combined with the developed mathematical tools, offers practical efficiency in determining critical control nodes in large probabilistic networks. The method is then applied to the human intracellular signal transduction network revealing that critical control nodes are associated with important biological features and perturbed sets of genes in human diseases, including SARS-CoV-2 target proteins and rare disorders. We believe that the proposed methodology can be useful to investigate multiple biological systems in which directed edges are probabilistic in nature, both in natural systems or when determined with large uncertainties in-silico.

Ligand-receptor systems, covalent modification cycles, and transcriptional networks are the fundamental components of cell signaling and gene expression systems. While their behavior in reaching a steady-state regime under step-like stimulation is well understood, their response under repetitive stimulation, particularly at early time stages is poorly characterized. Yet, early-stage responses to external inputs are arguably as informative as late-stage ones. In simple systems, a periodic stimulation elicits an initial transient response, followed by periodic behavior. Transient responses are relevant when the stimulation has a limited time span, or when the stimulated component's timescale is slow as compared to the timescales of the downstream processes, in which case the latter processes may be capturing only those transients. In this study, we analyze the frequency response of simple motifs at different time stages. We use dose-conserved pulsatile input signals and consider different metrics versus frequency curves. We show that in ligand-receptor systems, there is a frequency preference response in some specific metrics during the transient stages, which is not present in the periodic regime. We suggest this is a general system-level mechanism that cells may use to filter input signals that have consequences for higher order circuits. In addition, we evaluate how the described behavior in isolated motifs is reflected in similar types of responses in cascades and pathways of which they are a part. Our studies suggest that transient frequency preferences are important dynamic features of cell signaling and gene expression systems, which have been overlooked.

The strength of molecular interactions is characterized by their dissociation constants (K_D). Only high-affinity interactions (K_D ≤ 10^-8 M) are extensively investigated and support binary on/off switches. However, such analyses have discounted the presence of low-affinity binders (K_D > 10^-5 M) in the cellular environment. We assess the potential influence of low-affinity binders on high-affinity interactions. By employing Gillespie stochastic simulations and continuous methods, we demonstrate that the presence of low-affinity binders can alter the kinetics and the steady state of high-affinity interactions. We refer to this effect as 'herd regulation' and have evaluated its possible impact in two different contexts including sex determination in Drosophila melanogaster and in signalling systems that employ molecular thresholds. We have also suggested experiments to validate herd regulation in vitro. We speculate that low-affinity binders are prevalent in biological contexts where the outcomes depend on molecular thresholds impacting homoeostatic regulation.

Remarkable advances are being made in developing interventions for eliciting long-term remission of HIV-1 infection. The success of these interventions will obviate the need for lifelong antiretroviral therapy, the current standard-of-care, and benefit the millions living today with HIV-1. Mathematical modelling has made significant contributions to these efforts. It has helped elucidate the possible mechanistic origins of natural and post-treatment control, deduced potential pathways of the loss of such control, quantified the effects of interventions, and developed frameworks for their rational optimization. Yet, several important questions remain, posing challenges to the translation of these promising interventions. Here, we survey the recent advances in the mathematical modelling of HIV-1 control and remission, highlight their contributions, and discuss potential avenues for future developments.

We demonstrate that the assembly pathway method underlying assembly theory (AT) is an encoding scheme widely used by popular statistical compression algorithms. We show that in all cases (synthetic or natural) AT performs similarly to other simple coding schemes and underperforms compared to system-related indexes based upon algorithmic probability that take into account statistical repetitions but also the likelihood of other computable patterns. Our results imply that the assembly index does not offer substantial improvements over existing methods, including traditional statistical ones, and imply that the separation between living and non-living compounds following these methods has been reported before.

Integrating multi-omics data into predictive models has the potential to enhance accuracy, which is essential for precision medicine. In this study, we developed interpretable predictive models for multi-omics data by employing neural networks informed by prior biological knowledge, referred to as visible networks. These neural networks offer insights into the decision-making process and can unveil novel perspectives on the underlying biological mechanisms associated with traits and complex diseases. We tested the performance, interpretability and generalizability for inferring smoking status, subject age and LDL levels using genome-wide RNA expression and CpG methylation data from the blood of the BIOS consortium (four population cohorts, N_total = 2940). In a cohort-wise cross-validation setting, the consistency of the diagnostic performance and interpretation was assessed. Performance was consistently high for predicting smoking status with an overall mean AUC of 0.95 (95% CI: 0.90-1.00) and interpretation revealed the involvement of well-replicated genes such as AHRR, GPR15 and LRRN3. LDL-level predictions were only generalized in a single cohort with an R² of 0.07 (95% CI: 0.05-0.08). Age was inferred with a mean error of 5.16 (95% CI: 3.97-6.35) years with the genes COL11A2, AFAP1, OTUD7A, PTPRN2, ADARB2 and CD34 consistently predictive. For both regression tasks, we found that using multi-omics networks improved performance, stability and generalizability compared to interpretable single omic networks. We believe that visible neural networks have great potential for multi-omics analysis; they combine multi-omic data elegantly, are interpretable, and generalize well to data from different cohorts.

A complex interplay between various processes underlies the neuropathology of Alzheimer's disease (AD) and its progressive course. Several lines of evidence point to the coupling between Aβ aggregation and neuroinflammation and its role in maintaining brain homeostasis during the long prodromal phase of AD. Little is however known about how this protective mechanism fails and as a result, an irreversible and progressive transition to clinical AD occurs. Here, we introduce a minimal model of a coupled system of Aβ aggregation and inflammation, numerically simulate its dynamical behavior, and analyze its bifurcation properties. The introduced model represents the following events: generation of Aβ monomers, aggregation of Aβ monomers into oligomers and fibrils, induction of inflammation by Aβ aggregates, and clearance of various Aβ species. Crucially, the rates of Aβ generation and clearance are modulated by inflammation level following a Hill-type response function. Despite its relative simplicity, the model exhibits enormously rich dynamics ranging from overdamped kinetics to sustained oscillations. We then specify the region of inflammation- and coupling-related parameters space where a transition to oscillatory dynamics occurs and demonstrate how changes in Aβ aggregation parameters could shift this oscillatory region in parameter space. Our results reveal the propensity of coupled Aβ aggregation-inflammation systems to oscillatory dynamics and propose prolonged sustained oscillations and their consequent immune system exhaustion as a potential mechanism underlying the transition to a more progressive phase of amyloid pathology in AD. The implications of our results in regard to early diagnosis of AD and anti-AD drug development are discussed.