MEIGO

MEIGO implements metaheuristic optimization algorithms for parameter estimation, model fitting, continuous nonlinear programming (cNLP), mixed-integer programming (MINLP), and integer programming (IP) problems in systems biology and bioinformatics.

Key Features:

• Metaheuristic Algorithms: Implements enhanced scatter search (eSS) and variable neighborhood search (VNS), with eSS applied to cNLP and MINLP and VNS applied to IP.
• Parallel Processing: eSS and VNS can execute single-threaded or in parallel using a cooperative strategy to scale computations.
• Bayesian Inference (BayesFit): The R implementation includes BayesFit to perform parameter estimation via Bayesian inference.
• Implementations: Provided as R and Matlab toolboxes with Python accessibility via an R wrapper.
• Modular Structure: Modular design allows integration of additional optimization methods.

Scientific Applications:

• Parameter Estimation: Facilitates parameter estimation of biological models, including Bayesian parameter estimation via BayesFit.
• Model Fitting: Applied to model fitting and complex system modeling in systems biology and bioinformatics.
• Benchmark Evaluation: Has been evaluated against optimization benchmarks and applied across diverse case studies in bioinformatics and systems biology.

Methodology:

Implements enhanced scatter search (eSS) for cNLP and MINLP, variable neighborhood search (VNS) for IP, supports single-threaded and cooperative parallel execution, and includes BayesFit for Bayesian parameter estimation.