GNU MCSim

GNU MCSim performs Bayesian statistical inference of parameter values for systems biology models, including non-linear differential equation systems.

Key Features:

• Bayesian framework: Employs Bayesian numerical techniques to integrate prior information and multiple levels of uncertainty in parameter inference.
• Markov Chain Monte Carlo (MCMC) simulations: Uses MCMC to generate samples from the joint posterior distribution of model parameters given data and specified priors.
• Support for Systems Biology Markup Language (SBML): Starting with version 5.3.0, reads and simulates models encoded in SBML.
• Hierarchical statistical models: Supports hierarchical modeling to represent and infer multiple levels of variability in biological data.
• Optimal design of experiments: Facilitates investigation of optimal experimental designs to maximize information gain from data.
• Cross-platform implementation: Implemented in standard C and runs on platforms with a compatible C compiler.

Scientific Applications:

• Parameter estimation in systems biology: Estimation of mechanistic model parameters for models described by non-linear differential equations.
• Uncertainty quantification and model validation: Quantifies parameter uncertainty and supports model validation against experimental data via posterior sampling.
• SBML model simulation: Simulation and analysis of models encoded in Systems Biology Markup Language.
• Experimental design: Exploration of experimental designs to improve identifiability and information content of collected data.
• Hierarchical data analysis: Analysis of data with nested or multi-level structure using hierarchical statistical models.

Methodology:

Applies Bayesian numerical techniques and MCMC sampling from the joint posterior distribution, supports reading and simulation of SBML models, implements hierarchical statistical modeling, and enables investigation of optimal experimental designs.