For decades, clinical nutrition has worked from a starting assumption that is quietly wrong: that the same diet should produce roughly the same outcome in the same body. Two patients walk into the same clinic. Both follow the same Mediterranean-style protocol, both hit the macros, both stay compliant for twelve weeks. One sheds inflammation and ten kilos; the other plateaus, feels worse, and quietly disengages. The old explanation reached for willpower or adherence. The newer one — and the better one — is that their genes are not the same.
What nutrigenomics actually means
Nutrigenomics studies how nutrients and dietary patterns interact with the genome — both how food influences gene expression (sometimes called nutritional epigenetics) and how individual genetic variation changes the way the body handles specific nutrients. It overlaps with, but is not identical to, nutrigenetics, which focuses on how specific gene variants alter nutritional needs. In clinical practice the two are usually rolled together; the distinction matters more to researchers than to the patient sitting across the desk.
The mechanism, in plain terms, is that single-letter changes in DNA (called SNPs — single nucleotide polymorphisms) often change how efficiently an enzyme works, how a receptor binds, or how a transporter moves a molecule across a membrane. A person might convert dietary folate to its active form at half the rate of their neighbour, absorb fat-soluble vitamins more aggressively, or respond to refined carbohydrate with a sharper insulin spike. None of those differences are visible without testing. All of them shape what “optimal” actually looks like for that individual.
The clinical implication is straightforward: when a generic protocol fails repeatedly on a patient who seems to be doing everything right, the genome is increasingly the first place to look — not the last.
Five genes every practitioner should know
You do not need to memorise hundreds of SNPs to use nutrigenomics in practice. A handful of well-studied variants explain a disproportionate share of the inter-individual differences clinicians actually encounter.
MTHFR — folate metabolism
MTHFR encodes the enzyme that converts folate into its bioactive form, 5-MTHF. Carriers of the common C677T variant produce an enzyme that works at 30–70% of normal efficiency, leaving them functionally folate-deficient even on textbook intakes. Clinically this matters for cardiovascular risk, methylation status, neural tube defects in pregnancy, and mood. The plan changes meaningfully: switching from folic acid to methylfolate, scaling intake, and tracking homocysteine.
APOE — fat and cholesterol response
APOE comes in three main variants (ε2, ε3, ε4) that strongly influence how the body handles dietary saturated fat and cholesterol. ε4 carriers tend to mount sharper LDL responses to saturated fat and carry elevated cardiovascular and cognitive risk. A diet that is neutral for an ε3/ε3 patient can be actively harmful for an ε4 carrier. The plan changes: lower saturated fat, leaner protein, more omega-3, tighter lipid monitoring.
FTO — appetite regulation and weight
FTO’s common A-allele variants are associated with elevated BMI, larger appetite, and stronger preference for energy-dense foods. The encouraging news from the literature is that FTO-associated weight gain is more responsive to structured intervention than the average — these patients benefit disproportionately from protein-forward, satiety-driven meal patterns. The plan changes: protein at every meal, structured eating windows, less reliance on willpower alone.
TCF7L2 — carbohydrate sensitivity
TCF7L2 variants are among the strongest known genetic predictors of type 2 diabetes risk and influence insulin secretion in response to glucose. Carriers do not tolerate refined carbohydrate the way non-carriers do — postprandial spikes are sharper and pancreatic recovery is slower. The plan changes: lower glycaemic load, more fibre and protein, attention to meal sequencing, earlier consideration of continuous glucose data.
PEMT — choline requirements
PEMT encodes an enzyme that produces phosphatidylcholine endogenously. Variants reduce that capacity, raising the dietary choline requirement — important for liver health, methylation, and pregnancy. Choline-deficient diets can produce fatty liver and elevated liver enzymes in PEMT variant carriers even at population-average intakes. The plan changes: deliberately increased choline-rich foods (eggs, soy, cruciferous vegetables, beans, peanuts) and, in some cases, supplementation.
From DNA report to diet plan — the clinical translation gap
A DNA panel is the input, not the output. The most common failure mode in nutrigenomic practice is the same one that plagues laboratory medicine generally: a sophisticated test report handed to a patient with little or no interpretation. A list of SNPs without context becomes either a source of anxiety or a curiosity discarded within a week.
Translating DNA data into a plan requires integration. Genetic variants explain a meaningful share of inter-individual difference but rarely the majority. Lifestyle, gut microbiome composition, current bloodwork, life stage, and clinical goals all weigh in. A plan built on DNA alone will miss the patient who carries low-risk variants but presents with high inflammation; equally, it will miss the high-risk carrier who has compensated brilliantly through habit. The practitioner’s job is to layer the genetic signal onto the rest of the picture and design accordingly.
Where nutrigenomics is overhyped (and where it is just right)
It is worth being honest about what the field can and cannot do. Direct-to-consumer DNA reports often promise personalised meal plans built on dozens of “wellness” SNPs whose clinical evidence is thin to non-existent. Genes for “ideal coffee intake” or “cilantro preference” make for good marketing copy; they do not belong in a serious clinical protocol.
The robust applications — methylation status, fat metabolism, carb sensitivity, choline needs, vitamin D pathway, alcohol metabolism — are well-replicated in the literature and carry clear nutritional implications that change patient outcomes. Used with judgement, nutrigenomics is one of the most powerful additions a personalised practice can make. Used credulously, it becomes another wellness fad. The difference is interpretation.
A DNA report is a starting line, not a destination. It removes a layer of trial-and-error and lets a practitioner design with intent — but only when paired with sound clinical judgement, integration with the rest of the patient’s profile, and a willingness to revisit the plan as results come in. At Horta Nutrition, we integrate DNA insights with microbiome data, blood work, and lifestyle context to build personalised plans for practitioners in our network — so the protocol fits the person, not the other way around. If you want a deeper walkthrough of how DNA-based planning works in clinical practice, the practitioner’s guide to DNA-based nutrition is the place to start.
