Skip to main content

Table 2 Missing heritability and the limitations of genome wide and candidate gene association studies

From: Propelling the paradigm shift from reductionism to systems nutrition

Limitation

Comments

References

Epistatic Interactions

Association studies analyze a single variable (e.g., SNP) with a trait. GWAS correct each SNP for multiple comparisons. Well documented in animal models with increasing numbers of examples in humans. Accounting for interactions decreased the amount of missing heritability. New analytical methods are being developed to test for interactions.

[11, 61, 62, 67, 105, 114]

Ascertainment bias

Many phenotypes such as type 2 diabetes or obesity were poorly characterized. For example, analysis of NHANES data demonstrated that body mass index was poorly associated with markers of cardiometabolic health. Not limited to GWAS

[97]

Gene–environment interactions

All organisms have genetic variation, producing phenotypic variation in response to environmental factors—this is the basis of natural selection. High-density genotyping, exome, and whole-genome sequencing have proved that each genome differs from all others. Adaptation to local environments has produced selection of gene variants—e.g., lactase persistence in Europe, Africa, and part of the Mideast and selection for metabolizing high-fat diets in Greenland Inuits. Experimental systems have demonstrated gene–diet interactions but as with SNP–disease studies, the effect size is small.

[24, 38, 39, 57, 71, 109, 113]

Epigenetics

An epigenetic trait is a stably heritable phenotype resulting from changes in a chromosome without alterations in the DNA sequence. Although not measured in most association studies, DNA methylation and chromatin modifications alter the expression of genetic information differentially in each tissue. Parent-of-origin genomic imprinting also alters gene regulation.

[14, 29, 88, 112]