Unlocking data sets by calibrating populations of models to data density: A study in atrial electrophysiology

Lawson, Brodie; Drovandi, Christopher C.; Cusimano, Nicole; Burrage, Pamela; Rodriguez, Blanca; Burrage, Kevin

doi:10.1126/sciadv.1701676

Cited by 68 publications

(103 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Calibrating a model to the synthetic data Models were fitted to the observed noisy data using Bayesian inference (Girolami 2008), implemented by a posterior simulation method called Sequential Monte Carlo (SMC) sampling (Doucet et al 2000), a class of methods which is related to and overlaps with particle filtering (Doucet & Johansen 2009). SMC sampling is seeing increasing usage for a diverse range of applications (Drovandi & Pettitt 2011;Jeremiah et al 2012;Lawson et al 2018;Sisson et al 2018). Our SMC sampler (Appendix S2) is adapted from ideas presented in Jeremiah et al (2012) andDel Moral et al (2012).…”

Section: Generating the Simulated Datamentioning

confidence: 99%

Informing management decisions for ecological networks, using dynamic models calibrated to noisy time‐series data

et al. 2020

View full text Add to dashboard Cite

Well‐intentioned environmental management can backfire, causing unforeseen damage. To avoid this, managers and ecologists seek accurate predictions of the ecosystem‐wide impacts of interventions, given small and imprecise datasets, which is an incredibly difficult task. We generated and analysed thousands of ecosystem population time series to investigate whether fitted models can aid decision‐makers to select interventions. Using these time‐series data (sparse and noisy datasets drawn from deterministic Lotka‐Volterra systems with two to nine species, of known network structure), dynamic model forecasts of whether a species’ future population will be positively or negatively affected by rapid eradication of another species were correct > 70% of the time. Although 70% correct classifications is only slightly better than an uninformative prediction (50%), this classification accuracy can be feasibly improved by increasing monitoring accuracy and frequency. Our findings suggest that models may not need to produce well‐constrained predictions before they can inform decisions that improve environmental outcomes.

show abstract

Section: Generating the Simulated Datamentioning

confidence: 99%

Informing management decisions for ecological networks, using dynamic models calibrated to noisy time‐series data

et al. 2020

View full text Add to dashboard Cite

show abstract

“…On the other hand, spatial (cell-to-cell) variability has been suggested to be at least partly mediated by differential ionic contributions to the electrophysiology of individual cells [10], [11], [17], [18]. In this regard, the effect of variations in ion channel numbers to differences across cells has been well established, while that of variations in other characteristics related to ionic activation or inactivation is less clear [19], [20]. Computational modeling and simulation has greatly helped to shed light on the mechanisms underlying cardiac electrophysiological variability and its ability to predict arrhythmic risk in different settings [10], [11], [19].…”

mentioning

confidence: 99%

Data-Driven Identification of Stochastic Model Parameters and State Variables: Application to the Study of Cardiac Beat-to-Beat Variability

Sampedro-Puente

Fernandez-Bes

Virág

et al. 2020

IEEE J. Biomed. Health Inform.

View full text Add to dashboard Cite

Objective: Enhanced spatiotemporal ventricular repolarization variability has been associated with ventricular arrhythmias and sudden cardiac death, but the involved mechanisms remain elusive. In this work a methodology for estimation of parameters and state variables of stochastic human ventricular cell models from input voltage data is proposed for investigation of repolarization variability. Methods: The proposed methodology formulates state-space representations based on developed stochastic cell models and uses the Unscented Kalman Filter (UKF) to perform joint parameter and state estimation. Evaluation over synthetic and experimental data is presented. Results: Results on synthetically generated data show the ability of the methodology to: 1) filter out measurement noise from action potential (AP) traces; 2) identify model parameters and state variables from each of those individual AP traces, thus allowing robust characterization of cell-to-cell variability; and 3) replicate statistical population's distributions of input APbased markers, including dynamic markers quantifying beat-tobeat variability. Application onto experimental data demonstrates the ability of the methodology to match input AP traces while concomitantly inferring the characteristics of underlying stochastic cell models. Conclusion: A novel methodology is presented for estimation of parameters and hidden variables of stochastic cardiac computational models, with the advantage of providing a one-to-one match between each individual AP trace and a corresponding set of model characteristics. Significance: The proposed methodology can greatly help in the characterization of temporal (beat-to-beat) and spatial (cell-to-cell) variability in human ventricular repolarization and in ascertaining the corresponding underlying mechanisms, particularly in scenarios with limited available experimental data.

show abstract

“…This phenomenon has been 23 observed with a large number of viruses, mostly with RNA genomes [12,13]. Furthermore, it has been possible to demonstrate 24 that virions with defective genomes reduce the yield of virus from cells infected with wild type DENV and are known, 25 therefore, as defective interfering (DI) particles [14][15][16]. 26 There is an extensive literature on the activity of DI particles across a wide range of RNA viruses but interest waned in 27 the 1990s [13,17].…”

mentioning

confidence: 98%

“…We build an ensemble, population, of models, in 35 which each element in the population is a mathematical model with exactly the same framework, but where each model has a 36 different set of parameter values for the same set of parameters. All of these values are calibrated in some appropriate way 37 against multiple data [25]. In particular, we calibrate the data for plasma viral load and antibody response for 207 patients in 38 our population of models (POMs).…”

mentioning

confidence: 99%

See 1 more Smart Citation

A population of bang-bang switches of defective interfering particles makes within-host dynamics of dengue virus controllable

Mapder

Clifford

Aaskov

et al. 2018

Preprint

Self Cite

View full text Add to dashboard Cite

The titre of virus in a dengue patient and the duration of this viraemia has a profound effect on whether or not a mosquito will become infected when it feeds on the patient and this, in turn is a key driver of the magnitude of a dengue outbreak. As mosquitoes require 100-1000 times more virus to become infected than a patient, the transmission of dengue virus from a patient to a mosquito is a vulnerability that may be able to be targeted to improve disease control. The intrinsic variability in the within-host dynamics of viraemias is explored for a population of patients using the method of population of models (POMs). A dataset from 207 patients is used to calibrate 20,000 models for the infection kinetics for each of the four dengue virus serotypes. The effect of adding defective dengue virus interfering particles to patients as a therapeutic is evaluated using the calibrated POMs in a bang-bang optimal control setting.

show abstract

Unlocking data sets by calibrating populations of models to data density: A study in atrial electrophysiology

Abstract: We describe a statistically informed calibration of in silico populations to explore variability in complex systems.

Cited by 68 publications

References 71 publications

Informing management decisions for ecological networks, using dynamic models calibrated to noisy time‐series data

Informing management decisions for ecological networks, using dynamic models calibrated to noisy time‐series data

Data-Driven Identification of Stochastic Model Parameters and State Variables: Application to the Study of Cardiac Beat-to-Beat Variability

A population of bang-bang switches of defective interfering particles makes within-host dynamics of dengue virus controllable

Contact Info

Product

Resources

About