In July 2025, the Indian Journal of Clinical Biochemistry published research detailing the use of whole exome sequencing to predict type 2 diabetes (T2D), identifying ten single nucleotide polymorphisms (SNPs) linked to the disease, accounting for 90.5% of its variability. The study builds on a growing, and increasingly controversial, field: the use of polygenic risk scores (PRS) to assess an individual’s predisposition to a range of conditions, and, potentially, to select embryos with lower risk profiles.
The development of PRS relies on genome-wide association studies (GWAS), which identify genetic variants associated with specific traits. A recent study published in Nature, analyzing data from over 244,000 cases and 637,000 controls, found that the performance of T2D PRS is “context dependent,” performing better in younger, male individuals without hypertension, and who are not obese or overweight. The research too highlighted associations between PRS and diabetes-related complications like cardiometabolic traits and diabetic retinopathy.
While the potential for early risk assessment and preventative strategies is lauded by researchers – a study in Taiwan, detailed in Nature, achieved 84.2% accuracy in predicting T2D risk using a PRS model – concerns are mounting regarding the ethical and societal implications of applying this technology to reproductive choices. Genomic Prediction, a company offering genetic scores for various traits since 2020, discontinued advertising scores for “idiopathic short stature” and intellectual disability due to public controversy, though not, according to reports, because of fundamental scientific unreliability.
The accuracy of these scores remains a significant limitation. The theoretical maximum impact of a polygenic score for height, for example, is estimated at 2.5 inches, a figure not yet observed in studies. Selecting for multiple traits simultaneously reduces the accuracy of each individual prediction. The complex interplay of genes, where a single gene can influence multiple biological processes, introduces the risk of unintended consequences. A gene linked to an undesirable trait might also impact desirable ones in unforeseen ways.
Critics also point to the limited impact of genetic predisposition compared to environmental factors. A reduced risk of heart disease achieved through genetic screening, as demonstrated by the first couple to utilize the technology, was less than 1 percent – a difference potentially outweighed by lifestyle choices like diet and exercise. The focus on genetic selection, some argue, diverts attention from more readily achievable and equitable preventative measures.
A more profound concern centers on the potential for exacerbating social inequalities. While interventions like growth hormone injections and tutoring are accessible only to some, embryonic selection based on PRS could create a new class of “optimized” individuals, their status determined not by inherent capabilities but by their parents’ financial capacity to manipulate their genomes. This selection process, unlike individual interventions, impacts not only the selected individual but all of their descendants, potentially solidifying a genetic upper class.
Recent research, published in Nature, underscores the need to consider context when evaluating PRS, and acknowledges the potentially generalizable associations of T2D PRS with related traits, despite varying predictive performance across diverse populations. The study provides a comprehensive resource for characterizing T2D PRS, but does not address the long-term societal consequences of its application to reproductive technologies.
