The Grimaldi Framework

A comprehensive guide to the framework used by NutriGenAI to assess the scientific validity of nutrigenetic gene × diet interactions.

Overview

The Grimaldi et al. (2017) framework was developed within the EU FP7 Food4Me project to establish minimum standards of evidence for the scientific validity of nutrigenetic gene × diet interactions.

Published in Genes & Nutrition (DOI: 10.1186/s12263-017-0584-0), it was created by a consortium of experts including Keith A. Grimaldi, Ben van Ommen, Jose M. Ordovas, and others from the Global Nutrigenetics Knowledge Network.

The framework assesses four core dimensions to determine whether a gene × diet interaction has sufficient evidence to be considered scientifically valid for personalised dietary advice.

Criterion 1: Study Design & Quality

Studies are evaluated across seven sub-criteria:

•Interventional or observational design

•Prospective vs retrospective approach

•Randomised, placebo-controlled, blinded design

•Study power (subject numbers with effect allele)

•Effect magnitude

•P-values, FDR, and multiple testing corrections

•Replication in different populations and meta-analysis

Additionally: STREGA guideline compliance and quantitative dietary intake reporting are required.

Criterion 2: Type of Gene × Diet Interaction

DirectStrongest

Mechanistic interaction between the genetic variant and dietary component directly on a health biomarker. Example: MTHFR × folate → homocysteine levels.

IntermediateModerate

Mechanistic interaction, but other processes also affect the biomarker level. Example: MTHFR × riboflavin → blood pressure.

IndirectWeakest

Interaction influenced by gene × diet but affected by many other processes with long latency. Example: MTHFR × folate → CVD.

Criterion 3: Nature of the Genetic Variant

a) CausalDemonstrated causal effect on gene product function (e.g., enzyme activity or protein abundance).

b) In LD with Functional VariantVariant in linkage disequilibrium with a known functional variant (tagging SNP).

c) Associated, Unknown FunctionOnly statistical association; mechanism uncharacterised.

Criterion 4: Biological Plausibility

Biological plausibility is a judgement based on the collected evidence of a gene × diet interaction on a phenotype. It is rated as:

High

Medium

Low

Unknown

Overall Scientific Validity

The overall verdict synthesises all four criteria into a probability-based assessment:

Convincing>90%Multiple strong studies, fully explained mechanism

Probable66–90%Several studies, partly explained mechanism

Possible33–66%Few studies, partly explained mechanism

Not Demonstrated<33%Insufficient evidence

Read the original paper