Our main goal at Nell is to provide recommendations based on the most validated evidence to optimise health by analysing an individual’s genetic makeup. Our Science Advisory Board Team is dedicated to assessing scientific studies focused on personalised nutrition.
Each paper providing evidence for each polymorphism is assessed according to the Q-genie framework and only those polymorphisms with highest quality evidence shown across multiple studies are included in our report.
Quality of evidence
Scientific research follows a hierarchy of quality as follows:
- Meta-analysis and Systematic Review
- Genome Wide Associations Studies
- Randomized Controlled Trials
Meta-analysis / Systematic Reviews
High-quality meta-analysis and systematic reviews are one of the main sources Nell uses to validate science information. Both are used to critically assess relevant studies considering individual outcomes from different studies in an unbiased manner. In this way, we gathered a large amount of information, recognizing all the gaps as well as the useful benefit or harmful intervention for our customers. This is the best form to validate evidence and considered the top of the hierarchy of evidence.
GWAS provide reliable evidence of the susceptibility of different genetic variants to specific diseases. We can use this information alongside knowledge of an individual’s genetics to inform clinical practice and recommendations.
Genome Wide Association Studies
Another approach included as part of Nells’ Science Standards is the inclusion of Genome Wide Association Studies (GWAS). The aim of GWAS is to identify genotype-association across different genetic variants tested in multiple individuals.
Randomized and Observational studies
When meta-analysis or GWAS are not available for certain genetic variants, Nell aims to use research conducted in randomized controlled trials (RCT). RCT analyses the effectiveness of an intervention compared with a control group. It is considered a rigorous tool that explores the cause-effect associations between treatment or intervention and outcome. If RCT is not available, the next tier of evidence are observational studies with a good G-genie score.
In assessing the quality of the evidence we use the following key tools and concepts
- Hardy Weinberg Equilibrium
- Study Power
- Genotype errors and missing data
- Haplotype variation
The Q-genie framework is a well validated tool used to assess the reliability of genetic association studies. This tool assesses 11 dimensions including the rationale of the study, establishment of comparison groups (i.e. controls groups), technical and non-technical classification of the genetic variant analyzed, disclosure of source of bias, statistical power and sample size, statistical methods used, inclusion of Hardy-Weinberg equilibrium, interpretation of the results including all assumptions and interferences. Each question is marked on a 7-point Likert scale to give a final score: poor quality, moderate quality and good quality studies. At Nell only studies with good or moderate quality are included in the analysis.
Hardy Weinberg Equilibrium
It is recommended that nutrigenomic research should match the distribution of genotype frequencies in study samples with those the population studied. The estimated number of homozygous and heterozygous variant carriers is assessed through Hardy-Weinberg Equilibrium (HWE). Research with significant deviation from HWE can result in false positive results and impair the quality of the study. At Nell, we aim to include studies that utilize and are in line with HWE.
Study Power \ Sample size
Appropriate sample size is a very important aspect of a research in order to ensure sufficient power to detect the desired outcome. Any research should provide details about the sample size calculation when publishing. Without a power calculation is difficult to interpret and draw conclusions with the results obtained. Studies with underpowered sample size indicates absence of evidence. For these reasons, at Nell only randomized studies with details of statistical power are considered for analysis.
Genotype errors and reporting data
Transparency in reporting of genotyping protocol along with error rates is important in the detection of associations or linkage. The criteria for nutrigenomic studies that indicates a good genotyping quality are those with call rates above 95%.
Nature of genetic variant and modelling haplotype variation
In order to draw the conclusion for Nell’s report, we consider all haplotypes or Linkage Disequilibrium (LD) for each genetic variant. A haplotype is a combination of certain alleles that are neighbour genes and may be inherited together. The consideration of haplotype aims to test different numbers of SNPs to check their effectiveness with the outcome than just one single variant. Further, it improves the understanding of possible effects of specific genetic variants as well as enabling comparison of results across several studies.