Identifying Biochemical Reaction Networks From Heterogeneous Datasets

in IEEE Conference on Decision and Control, Osaka, Japan, December 2015 (2015, December)

Critical transitions in chronic disease: transferring concepts from ecology to systems medicine

in Current Opinion in Biotechnology (2015), 34

Ecosystems and biological systems are known to be inherently complex and to exhibit nonlinear dynamics. Diseases such as microbiome dysregulation or depression can be seen as complex systems as well and ... [more ▼]

Ecosystems and biological systems are known to be inherently complex and to exhibit nonlinear dynamics. Diseases such as microbiome dysregulation or depression can be seen as complex systems as well and were shown to exhibit patterns of nonlinearity in their response to perturbations. These nonlinearities can be revealed by a sudden shift in system states, for instance from health to disease. The identification and characterization of early warning signals which could predict upcoming critical transitions is of primordial interest as prevention of disease onset is a major aim in health care. In this review, we focus on recent evidence for critical transitions in diseases and discuss the potential of such studies for therapeutic applications. [less ▲]

On minimal realisations of dynamical structure functions

in Automatica (2015), 55

Motivated by the fact that transfer functions do not contain structural information about networks, dynamical structure functions were introduced to capture causal relationships between measured nodes in ... [more ▼]

Motivated by the fact that transfer functions do not contain structural information about networks, dynamical structure functions were introduced to capture causal relationships between measured nodes in networks. From the dynamical structure functions, a) we show that the actual number of hidden states can be larger than the number of hidden states estimated from the corresponding transfer function; b) we can obtain partial information about the true state-space equation, which cannot in general be obtained from the transfer function. Based on these properties, this paper proposes algorithms to find minimal realisations for a given dynamical structure function. This helps to estimate the minimal number of hidden states, to better understand the complexity of the network, and to identify potential targets for new measurements. [less ▲]

Transitively Consistent and Unbiased Multi-Image Registration Using Numerically Stable Transformation Synchronisation

in MIDAS Journal (2015)

Abstract. Transitive consistency of pairwise transformations is a desir- able property of groupwise image registration procedures. The transfor- mation synchronisation method [4] is able to retrieve ... [more ▼]

Abstract. Transitive consistency of pairwise transformations is a desir- able property of groupwise image registration procedures. The transfor- mation synchronisation method [4] is able to retrieve transitively con- sistent pairwise transformations from pairwise transformations that are initially not transitively consistent. In the present paper, we present a numerically stable implementation of the transformation synchronisa- tion method for a ne transformations, which can deal with very large translations, such as those occurring in medical images where the coor- dinate origins may be far away from each other. By using this method in conjunction with any pairwise (a ne) image registration algorithm, a transitively consistent and unbiased groupwise image registration can be achieved. Experiments involving the average template generation from 3D brain images demonstrate that the method is more robust with re- spect to outliers and achieves higher registration accuracy compared to reference-based registration. [less ▲]

Modelling the natural history of Huntingtons disease progression

in BMJ Open (2014)

Background: The lack of reliable biomarkers to track disease progression is a major problem in clinical research of chronic neurological disorders. Using Huntington’s disease (HD) as an example, we ... [more ▼]

Background: The lack of reliable biomarkers to track disease progression is a major problem in clinical research of chronic neurological disorders. Using Huntington’s disease (HD) as an example, we describe a novel approach to model HD and show that the progression of a neurological disorder can be predicted for individual patients. Methods : Starting with an initial cohort of 343 patients with HD that we have followed since 1995, we used data from 68 patients that satisfied our filtering criteria to model disease progression, based on the Unified Huntington’s Disease Rating Scale (UHDRS), a measure that is routinely used in HD clinics worldwide. Results : Our model was validated by: (A) extrapolating our equation to model the age of disease onset, (B) testing it on a second patient data set by loosening our filtering criteria, (C) cross-validating with a repeated random subsampling approach and (D) holdout validating with the latest clinical assessment data from the same cohort of patients. With UHDRS scores from the past four clinical visits (over a minimum span of 2 years), our model predicts disease progression of individual patients over the next 2 years with an accuracy of 89–91%. We have also provided evidence that patients with similar baseline clinical profiles can exhibit very different trajectories of disease progression. Conclusions : This new model therefore has important implications for HD research, most obviously in the development of potential disease-modifying therapies. We believe that a similar approach can also be adapted to model disease progression in other chronic neurological disorders. [less ▲]

Network Reconstruction from Intrinsic Noise: Non-Minimum-Phase Systems

in The proceedings of the The 19th World Congress of The International Federation of Automatic Control (2014)

This paper considers the problem of inferring the structure and dynamics of an unknown network driven by unknown noise inputs. Equivalently we seek to identify direct causal dependencies among manifest ... [more ▼]

This paper considers the problem of inferring the structure and dynamics of an unknown network driven by unknown noise inputs. Equivalently we seek to identify direct causal dependencies among manifest variables only from observations of these variables. We consider linear, time-invariant systems of minimal order and with one noise source per manifest state. It is known that if the transfer matrix from the inputs to manifest states is minimum phase, then this problem has a unique solution, irrespective of the network topology. Here we consider the general case where the transfer matrix may be non-minimum phase and show that solutions are characterized by an Algebraic Riccati Equation (ARE). Each solution to the ARE corresponds to at most one spectral factor of the output spectral density that satisfies the assumptions made. Hence in general the problem may not have a unique solution, but all solutions can be computed by solving an ARE and their number may be finite. [less ▲]

Understanding and Predicting Biological Networks Using Linear System Identi cation

in Kulkarni, V.; Stan, G.; Raman, K. (Eds.) A Systems Theoretic Approach to Systems and Synthetic Biology I: Models and System Characterizations (2014)

Finite-time road grade computation for a vehicle platoon

in IEEE (2014)

Given a platoon of vehicles traveling uphill, this paper considers the finite-time road grade computation problem. We propose a decentralized algorithm for an arbitrarily chosen vehicle to compute the ... [more ▼]

Given a platoon of vehicles traveling uphill, this paper considers the finite-time road grade computation problem. We propose a decentralized algorithm for an arbitrarily chosen vehicle to compute the road grade in a finite number of time-steps by using only its own successive velocity measurements. Simulations then illustrate the theoretical results. These new results can be applied to real-world vehicle platooning problems to reduce fuel consumption and carbon dioxide emissions. [less ▲]

Biexcitability and Bursting Mechanisms in Neural and Genetic Circuits

in The proceedings of the 19th World Congress The International Federation of Automatic Control (2014)

This paper compares mechanisms for generating repetitive spikes (bursts) in neural and transcriptional circuits. Neurons generate bursts followed by refractory periods controlled by ion channels in the ... [more ▼]

This paper compares mechanisms for generating repetitive spikes (bursts) in neural and transcriptional circuits. Neurons generate bursts followed by refractory periods controlled by ion channels in the membrane. In contrast, in gene transcription the bursts occur during a short time period followed by silent periods regulated by sis-regulatory elements. The role of excitability in producing different patterns of bursts is discussed by comparing the topology of a neural model with natural and synthetic transcriptional genetic circuits. In particular, a special bi-excitable architecture which embeds two excitable states are compared in these systems. [less ▲]

Distributed Kalman Filter with minimum-time covariance computation

in The proceedings of the IEEE 52nd Annual Conference on Decision and Control (2013)

This paper considerably improves the well-known Distributed Kalman Filter (DKF) algorithm by Olfati-Saber (2007) by introducing a novel decentralised consensus value computation scheme, using only local ... [more ▼]

This paper considerably improves the well-known Distributed Kalman Filter (DKF) algorithm by Olfati-Saber (2007) by introducing a novel decentralised consensus value computation scheme, using only local observations of sensors. It has been shown that the state estimates obtained in [8] and [9] approaches those of the Central Kalman Filter (CKF) asymptotically. However, the convergence to the CKF can sometimes be too slow. This paper proposes an algorithm that enables every node in a sensor network to compute the global average consensus matrix of measurement noise covariance in minimum time without accessing global information. Compared with the algorithm in [8], our theoretical analysis and simulation results show that the new algorithm can offer improved performance in terms of time taken for the state estimates to converge to that of the CKF. [less ▲]