PEtab—Interoperable specification of parameter estimation problems in systems biology

  title={PEtab—Interoperable specification of parameter estimation problems in systems biology},
  author={Leonard Schmiester and Yannik Schalte and Frank T. Bergmann and Tacio Camba and Erika Dudkin and Janine Egert and Fabian Frohlich and Lara Fuhrmann and Adrian L Hauber and Svenja Kemmer and Polina Lakrisenko and Carolin Loos and S. A. M. Merkt and Dilan Pathirana and Elba Raim'undez and Lukas Refisch and Marcus Rosenblatt and Paul Stapor and Philip A. Stadter and Dantong Wang and Franz-Georg Wieland and Julio R. Banga and Jens Timmer and Alejandro Fern{\'a}ndez Villaverde and Sven Sahle and Clemens Kreutz and Jan Hasenauer and Daniel Weindl Institute of Computational Biology and Helmholtz Zentrum Munchen -- German Research Center for Envi Health and Center for Mathematics and Technische Universitat Munchen and BioQUANTCOS and Heidelberg Northwestern University and Department of Applied Mathematics and University of Vigo and BioProcess Engineering Group and IIM-CSIC and Faculty of Mathematics and Natural Sciences and University of Bonn and Faculty of Veterinary Medicine and Medical Center and Institute of Medical Biometry and Statistics and Fdm - Freiburg Center for Data Analysis and Modeling and University of Freiburg and Department of Biology and Harvard Medical School and Institute of Metal Physics and Ragon Institute of Mgh and Mit and Harvard and Cambridge and Department of Mechanical Engineering and Massachusetts Institute of Technology and Signalling Research Centres Bioss and Cibss},
  journal={PLoS Computational Biology},
Reproducibility and reusability of the results of data-based modeling studies are essential. Yet, there has been—so far—no broadly supported format for the specification of parameter estimation problems in systems biology. Here, we introduce PEtab, a format which facilitates the specification of parameter estimation problems using Systems Biology Markup Language (SBML) models and a set of tab-separated value files describing the observation model and experimental data as well as parameters to… 

Figures and Tables from this paper

Fides: Reliable trust-region optimization for parameter estimation of ordinary differential equation models

This work implemented an established trust-region method as a modular Python framework (fides) to enable systematic comparison of different approaches to ODE model calibration involving various Hessian approximation schemes and proposes a novel hybrid Hessian approximations scheme that enhances optimizer performance and outperforms existing hybrid approaches.

Efficient computation of adjoint sensitivities at steady-state in ODE models of biochemical reaction networks

A new gradient computation method is proposed that facilitates the parameterization of large-scale models based on steady-state measurements that can be combined with existing gradient computation methods for time-course measurements.

Computational Speed-Up of Large-Scale, Single-Cell Model Simulations Via a Fully-Integrated SBML-Based Format

Two solutions to communication bottlenecks between gene expression and protein biochemistry modules are developed that speed up simulation by ~4-fold for hybrid stochastic-deterministic simulations and by over 100- fold for fully deterministic simulations.

Combining hypothesis- and data-driven neuroscience modeling in FAIR workflows

It is argued that the same imperatives that drive resources and policy for data – such as the FAIR principles – also support the integration of different modeling approaches and allow researchers to find, reuse, question, validate, and extend published models, regardless of whether they are implemented phenomenologically or mechanistically.

Deciphering signal transduction networks in the liver by mechanistic mathematical modelling

Integration of signal transduction with metabolic regulation in the liver and mechanistic links to translational approaches promise to provide predictive tools for biology and personalized medicine.

Disentangling ERBB Signaling in Breast Cancer Subtypes—A Model-Based Analysis

A unified mathematical model is developed that can describe the ERBB receptor and downstream signaling in response to therapeutic drugs targeting this clinically relevant signaling network in cell line that represent three major subtypes of breast cancer.

Dynamic modeling of Nrf2 pathway activation in liver cells after toxicant exposure

The quantitative model can describe the activation of the Nrf2 pathway by compounds with a different mechanism of activation, including drugs known for their ability to cause drug induced liver-injury (DILI) i.e., diclofenac and omeprazole (OMZ).

Bayesian parameter estimation for dynamical models in systems biology

A comprehensive framework for Bayesian parameter estimation and complete quantification of the effects of uncertainties in the data and models is proposed and applied to a series of signaling models of increasing mathematical complexity to highlight how focused uncertainty quantification can enrich systems biology modeling and enable additional quantitative analyses for parameter estimation.

pyABC: Efficient and robust easy-to-use approximate Bayesian computation

This work implements algorithms to accurately account for measurement noise, to adaptively scale-normalize distance metrics, to robustly handle data outliers, to elucidate informative data points via regression models, to circumvent summary statistics via optimal transport based distances, and to avoid local optima in acceptance threshold sequences by predicting acceptance rate curves.

Model Integration in Computational Biology: The Role of Reproducibility, Credibility and Utility

It is argued that without addressing the reproducibility crisis, scientists will have diminished ability to build, disseminate, and implement high-impact multi-scale modeling that is needed to understand the health crises the authors face.



The systems biology markup language (SBML): a medium for representation and exchange of biochemical network models

This work summarizes the Systems Biology Markup Language (SBML) Level 1, a free, open, XML-based format for representing biochemical reaction networks, a software-independent language for describing models common to research in many areas of computational biology.

SBML2Julia: interfacing SBML with efficient nonlinear Julia modelling and solution tools for parameter optimization

SBML2Julia translates optimisation problems specified in SBML and TSV files into Julia for Mathematical Programming (JuMP), executes the optimization and returns the results in tabular format, and is freely available under the MIT license.

Parameterization of mechanistic models from qualitative data using an efficient optimal scaling approach

The optimal scaling method is considered, which has been developed in statistics for categorical data and has been applied to dynamical systems, and a reduced formulation is derived for the optimization problem defining the optimal scaling.

Bayesian inference using qualitative observations of underlying continuous variables

Remarkably, estimates of parameter values derived from the qualitative data were nearly as consistent with the assumed ground-truth parameter values as estimatesderived from the lower throughput quantitative data, providing further motivation for leveraging qualitative data in biological modeling.

Statistical inference of mechanistic models from qualitative data using an efficient optimal scaling approach

The optimal scaling method is considered, which has been developed in statistics for categorical data and has been applied to dynamical systems, and a reduced formulation is derived for the optimization problem defining the optimal scaling.

Dynamic Modeling, Parameter Estimation, and Uncertainty Analysis in R

The R package dMod is presented, providing a framework for dealing with the inverse problem in dynamic systems modeled by ordinary differential equations, and provides a variety of functions to set up ordinary differential equation models, observation functions and parameter transformations for multi-conditional parameter estimation.

Efficient parameterization of large-scale dynamic models based on relative measurements

A novel hierarchical approach combining the efficient analytic evaluation of optimal scaling, offset, and error model parameters with the scalable evaluation of objective function gradients using adjoint sensitivity analysis that allows for the efficient parameterization of large-scale models based on heterogeneous relative measurements.

Benchmark problems for dynamic modeling of intracellular processes

This work presents a collection of 20 ODE models developed given experimental data as benchmark problems in order to evaluate new and existing methodologies, e.g. for parameter estimation or uncertainty analysis.

Using both qualitative and quantitative data in parameter identification for systems biology models

The authors demonstrate an approach that allows use of both quantitative and qualitative data for parameterising dynamical models, and indicates the value of combining qualitative and quantitative data to parameterize systems biology models.