For decades, clinical biomarkers have been evaluated against population-derived reference intervals and fixed diagnostic thresholds. A laboratory value falling outside a defined range triggers suspicion of disease, regardless of who the patient is. Yet an expanding body of evidence demonstrates that genetic variation, epigenetic regulation, environmental exposures and their interactions can influence biomarker concentrations, disease susceptibility and therapeutic response. The challenge, the editorial argues, is shifting from a one-size-fits-all application of biomarkers toward understanding the biological context that determines what a given measurement means for a specific individual.

Among the most striking examples is a study by van der Laan and colleagues, who used Oxford Nanopore long-read sequencing in five individuals with Hao-Fountain syndrome. In those patients, the team identified pathogenic single-nucleotide variants in the USP7 gene, structural deletions and a characteristic DNA methylation episignature, a genome-wide methylation pattern associated with the syndrome. The significance lies in the integration: a single sequencing platform simultaneously captured genomic and epigenomic information that traditionally required separate assays. The authors suggest such approaches could eventually allow several marker types to be measured in one assay, potentially improving risk estimates while reducing cost.

The epigenetic dimension is developed further in a review by Munns and colleagues on type 2 diabetes epigenetic biomarkers. The review highlights the potential of DNA methylation signatures for risk prediction and monitoring in type 2 diabetes, while candidly identifying limitations in reproducibility, genome coverage and population representation. Crucially, the authors call for sequencing-based methylation profiling, larger longitudinal cohorts and the meaningful inclusion of genetically diverse and Indigenous populations. Their argument is pointed: technological innovation and health equity must progress together if precision medicine is to deliver broadly applicable benefits rather than benefits confined to well-studied populations.

Population context is equally evident in a preliminary study by Saed and colleagues, who examined polymorphisms in the CTLA-4 gene in Kurdish patients in Iraq. In a cohort of 52 patients with type 1 diabetes and 21 healthy controls, the G allele of the CTLA-4 +49 A>G variant (rs231775) was associated with increased susceptibility to type 1 diabetes, whereas the −318 C>T variant was not. CTLA-4 is an immune checkpoint receptor that acts as a brake on T-cell activation; these polymorphisms may influence susceptibility by altering the expression and cell-surface availability of CTLA-4, potentially permitting enhanced autoreactive T-cell-mediated destruction of insulin-producing pancreatic beta cells. The finding requires confirmation in larger cohorts, but the study illustrates the value of extending genetic association research to populations historically underrepresented in genomic databases.

Population-specific evaluation is not limited to DNA sequence. Bahijri and colleagues investigated glycated albumin as a marker of glycaemic control in 132 adults from Saudi Arabia, finding that it correlated with established measures of glycaemic status and distinguished normoglycaemia from intermediate hyperglycaemia and diabetes. Glycated albumin reflects shorter-term glycaemia than glycated haemoglobin and can be useful where haemoglobin-based measures are unreliable. The findings support further evaluation of glycated albumin as an alternative or complementary glycaemic biomarker, while underscoring that appropriate reference ranges must be established in the population in which the marker will actually be used.

Not every genetic association translates into clinically useful prediction, as two further contributions demonstrate. Ji and colleagues analysed UGT1A1 genotypes in 115 newborns undergoing genetic screening and found that biallelic or multi-allelic variants were associated with mildly higher transcutaneous bilirubin concentrations at six weeks, suggesting slower postnatal bilirubin clearance. However, clinically significant neonatal hyperbilirubinemia was uncommon and did not differ significantly between variant carriers and controls. The study shows that the path from a detectable genotype-phenotype association to prediction of clinically significant disease is not always straightforward.

Statistical caution is reinforced by Wang and colleagues, who performed an updated meta-analysis of 53 studies on the TNF-alpha -308G/A and -238G/A polymorphisms and systemic lupus erythematosus susceptibility. Although conventional meta-analysis identified several statistically significant associations, evaluation using prediction intervals and Bayesian false-discovery probability indicated that these were less-credible positive findings. The work is a timely reminder that statistical association alone is insufficient for biomarker discovery; reproducibility, effect size, heterogeneity, population structure and the probability of false-positive findings must all be weighed before a candidate marker earns clinical credibility.

Genomic information may also personalise treatment rather than diagnosis. A narrative review by Hamdani and colleagues summarised gene polymorphisms associated with adverse effects of the immunosuppressants methotrexate, azathioprine, cyclophosphamide and mycophenolate mofetil across 23 studies in systemic lupus erythematosus. The review illustrates how pharmacogenomic biomarkers could move precision medicine beyond disease susceptibility toward drug selection, dosage optimisation and avoidance of treatment toxicity. Meanwhile, Appleby and Purcell broadened the concept beyond the human genome entirely, reviewing circulating bacterial DNA as a potential diagnostic and prognostic marker in cardiovascular disease and examining microbial translocation as one possible biological source, though heterogeneity in study populations, sample types and analytical methods currently limits clinical translation.

Together, these eight contributions point to clear priorities for the next phase of biomarker discovery. The field is moving beyond single genetic variants and isolated measurements toward integrated, multi-modal markers combining genomic, epigenomic, proteomic, microbial, environmental, imaging and clinical information. Associations will require rigorous replication and demonstrated clinical utility, and reference intervals and diagnostic thresholds should be evaluated across diverse populations rather than assumed to be universally transferable. The editorial’s closing message is ambitious but clear: the goal of the genomics era is not simply to measure biomarkers more accurately, but to interpret them more meaningfully, linking genomic diversity to diagnosis, prognosis, therapeutic response and genuinely equitable precision medicine.

Subject of Research: Genomic and epigenomic advances in biomarker discovery and interpretation

Article Title: Editorial: Advancements in biomarker genetics: insights from the genomics era

Article References: Cree, S. L., Lassé, M., & Pilbrow, A. P. (2026). Editorial: Advancements in biomarker genetics: insights from the genomics era. Frontiers in Molecular Biosciences, 13, Article 2014518. https://doi.org/10.3389/fmolb.2026.2014518

Image Credits: AI Generated

DOI: 10.3389/fmolb.2026.2014518

Keywords: biomarkers, genomics, epigenetics, DNA methylation, nanopore sequencing, pharmacogenomics, population diversity, precision medicine, type 1 diabetes, systemic lupus erythematosus, glycated albumin, health equity