A framework to establish credibility of computational models in biology

Patterson, EA ORCID: 0000-0003-4397-2160 and Whelan, MP (2017) A framework to establish credibility of computational models in biology. Progress in Biophysics and Molecular Biology, 129. pp. 13-19.

Text credibility of biological models post review v1 clean.docx - Author Accepted Manuscript Download (187kB)

Abstract

Computational models in biology and biomedical science are often constructed to aid people's understanding of phenomena or to inform decisions with socioeconomic consequences. Model credibility is the willingness of people to trust a model's predictions and is often difficult to establish for computational biology models. A 3 × 3 matrix has been proposed to allow such models to be categorised with respect to their testability and epistemic foundation in order to guide the selection of an appropriate process of validation to supply evidence to establish credibility. Three approaches to validation are identified that can be deployed depending on whether a model is deemed untestable, testable or lies somewhere in between. In the latter two cases, the validation process involves the quantification of uncertainty which is a key output. The issues arising due to the complexity and inherent variability of biological systems are discussed and the creation of ‘digital twins’ proposed as a means to alleviate the issues and provide a more robust, transparent and traceable route to model credibility and acceptance.

Item Type:	Article
Uncontrolled Keywords:	Computational biology, Computational models, Credibility, Validation, In silico medicine, Biomedical modelling
Depositing User:	Symplectic Admin
Date Deposited:	02 Aug 2016 10:55
Last Modified:	19 Jan 2023 07:32
DOI:	10.1016/j.pbiomolbio.2016.08.007
Related URLs:	Author Publisher
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3002689