A team led by Herbert M. Sauro and colleagues from the Center for Reproducible Biomedical Modeling has published “From FAIR to CURE: guidelines for computational models of biological systems” in npj Systems Biology and Applications, a Nature Portfolio journal. Developed with collaborators from institutions around the world, the paper addresses how mechanistic computational models can be made more reliable, understandable, and useful to other researchers.
The publication introduces the CURE principles, which call for computational models to be Credible, Understandable, Reproducible, and Extensible. These principles complement the established FAIR guidelines for scientific data while addressing the distinct needs of models used in biology, medicine, physiology, and systems biology.
The paper explains how model credibility can be strengthened through verification, validation, and the evaluation of uncertainty. It also emphasizes clear descriptions and annotations, adherence to modeling standards and open science practices, and the use of formats and code that allow models to be reproduced, adapted, and reused.
The authors provide baseline and recommended requirements for applying the CURE principles, along with a suggested checklist for research-based models. This practical framework is intended to improve the trustworthiness and long-term value of computational models, particularly as their use expands in biomedical research and emerging applications such as digital twins.
Read the full article:
https://www.nature.com/articles/s41540-026-00651-0