Improving dynamic predictions with ensembles of observable models (supplementary information)

This repository contains the accompanying code to the paper "Improving dynamic predictions with ensembles of observable models", by Gemma Massonis, Alejandro F. Villaverde, and Julio R. Banga (gemma.massonis@csic.es, afvillaverde@uvigo.gal, j.r.banga@csic.es), which presents a method for building ensembles of dynamic models. 

The repository contains four folders, one for each case study considered in the paper. Each folder contains:

• A Matlab Live Script that analyses the case study, provided in two formats:
  • as .html files, for convenient visualization, and
  • as .mlx files that can be executed in Matlab.
• Auxiliary files including model definitions, simulated data, and others.

In order to run the Live Scripts, the following software is needed and must be in the Matlab path:

• MATLAB (tested with version R2020b), including the following toolboxes and packages:
  • Statistics Toolbox 
  • FanChart, from the Matlab file exchange (https://es.mathworks.com/matlabcentral/fileexchange/48006-fanchart-visualize-percentiles-of-time-series-data)
  • gramm (https://github.com/piermorel/gramm)
• AMICI (tested with version 0.11.12) (https://github.com/AMICI-dev/AMICI/releases/tag/v0.11.12)
• MEIGO (tested with version V18JAN2018) (https://github.com/gingproc-IIM-CSIC/MEIGO64)
• AMIGO2 (tested with AMIGO2_R2019b) (https://sites.google.com/site/amigo2toolbox/)