🧬 biology

Carrier Burden and Birth Prevalence of Recessive Inborn Errors of Metabolism in South Asians: A gnomAD and GenomeIndia Study

This study leverages gnomAD v4 and GenomeIndia data to estimate that South Asians have the highest collective carrier frequency (84.2%) for recessive inborn errors of metabolism among global ancestries, with a pooled birth prevalence of approximately 1 in 529, while also highlighting significant Eurocentric biases in existing ClinVar variant annotations.

Original authors: Atanu Dutta

Published 2026-10-20

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Atanu Dutta

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ⚕️ This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer

Every human carries a hidden library of genetic instructions, written in a code that determines how our bodies function. Occasionally, a single letter in this code is misspelled. When such a misspelling occurs in a gene responsible for breaking down food or processing chemicals, it can lead to a condition known as an inborn error of metabolism. These are rare disorders that disrupt the body's chemical pathways, often causing severe illness or developmental challenges early in life. Most of these conditions are recessive, meaning a child is only affected if they inherit two copies of the flawed instruction—one from each parent. If a person carries just one copy, they are usually healthy but can pass the error to their children. Because these errors are individually rare but collectively numerous, understanding how common they are in specific populations is vital for public health, yet the full picture has remained difficult to see in South Asia.

A new study by Atanu Dutta, a researcher at the All India Institute of Medical Sciences in Kalyani, has brought this hidden landscape into sharper focus. By analyzing genetic data from over 86,000 people of South Asian descent, the researcher calculated how many individuals carry these metabolic errors and how often they might appear in newborns. The study did not rely on clinical records of sick children, which often miss mild or undiagnosed cases. Instead, it looked directly at the genetic code itself, scanning nearly 700 genes known to be linked to these metabolic disorders. The researcher used a massive global database of genetic information, filtering it to find specific types of harmful changes, and then validated their findings against a separate, independent dataset of nearly 10,000 individuals from diverse communities across India.

The results reveal a surprisingly high burden of these genetic risks within the population. The study estimates that approximately 84 percent of South Asians carry at least one pathogenic allele, a term for a harmful genetic variant linked to these disorders. This means that the vast majority of people in this region are carriers for at least one of these conditions. When the researcher calculated the likelihood of two carriers having a child with the disease, assuming parents choose partners randomly without regard to family ties, they found that about 255 out of every 100,000 births would be affected. This translates to roughly one in every 392 babies. This figure is a composite of many different conditions; no single disease accounts for the majority of the risk. Instead, the burden is spread across dozens of genes, with the most common contributors including genes linked to conditions like Wilson disease, phenylketonuria, and various glycogen storage disorders.

However, the story changes significantly when the reality of family structure is taken into account. South Asia is home to thousands of endogamous communities, groups that traditionally marry within their own circle, and a significant number of marriages occur between close relatives, such as first cousins. This practice increases the chance that a child will inherit two copies of the same harmful gene. The researcher modeled different levels of this family closeness, ranging from low to high. In scenarios where parents are first cousins, the estimated rate of affected births jumps dramatically to 5,824 per 100,000, or one in every 17 children. Even in scenarios with lower levels of family intermarriage, the risk rises to between one in 86 and one in 34 births. These numbers suggest that while the genetic risk is widespread, the actual occurrence of disease is heavily influenced by local marriage customs.

The researcher also compared their findings with data from other global populations to ensure their methods were sound. They discovered that South Asians carry a collective burden of these metabolic errors that is among the highest in the world, comparable to or slightly higher than other large, diverse populations. This contrasts with earlier assumptions that might have suggested lower risks. The study also highlighted a bias in existing medical databases, which have historically focused heavily on European populations. When the researcher looked only at variants that had been previously confirmed by doctors in those older databases, the risk for South Asians appeared much lower, while the risk for Europeans appeared higher. This reversal suggested that many harmful variants in South Asian populations had simply been overlooked by the global medical community. By using advanced computer tools to predict which genetic changes are likely to be harmful, the study was able to uncover these hidden risks, providing a more accurate and equitable picture of genetic health.

To confirm their results, the researcher applied their calculations to a separate dataset from the GenomeIndia project, which includes individuals from 83 distinct endogamous groups. While this smaller group showed a slightly lower overall prevalence, the pattern of which genes were most common remained consistent, reinforcing the reliability of the main findings. The study also identified a few specific genetic variants that seemed to be unique to certain Indian communities, though some of these required further verification to ensure they were truly harmful and not just harmless variations. The researcher noted that their estimates might still be conservative, as current genetic testing methods can miss certain types of complex structural changes in the DNA. Even with these limitations, the work provides a crucial baseline for understanding the genetic health of South Asia.

The implications of these findings are practical and immediate. Because the risk is spread across so many different genes, standard screening panels designed for European populations may miss a significant number of carriers in South Asia. The study suggests that new screening programs need to be tailored specifically to the genetic makeup of South Asian communities, focusing on the specific genes that are most common in the region. Furthermore, the data underscores the importance of genetic counseling, particularly in communities where family intermarriage is common, to help families understand their risks and make informed decisions. By mapping the true extent of these metabolic errors, this research offers a clearer path toward preventing illness and improving health outcomes for millions of people.

Technical Summary: Carrier Burden and Birth Prevalence of Recessive Inborn Errors of Metabolism in South Asians

Problem Statement
Inborn errors of metabolism (IEMs) are individually rare but collectively significant causes of early-life illness, intellectual disability, and mortality, predominantly inherited in an autosomal recessive (AR) manner. Estimating the prevalence of these disorders in South Asia has been challenging due to heterogeneous study designs, limited screening coverage, and the exclusion of predicted pathogenic variants. Furthermore, existing estimates often fail to account for the region's high rates of endogamy and consanguinity, which increase autozygosity and the likelihood of homozygous recessive states. Previous models have suggested prevalence rates ranging from 1 in 1,000 to 1 in 3,100 live births, but these figures may be understated. This study addresses the need for a comprehensive, genetically grounded estimation of the carrier burden and birth prevalence of AR-IEMs across the South Asian (SAS) population, while validating findings against independent cohort data and correcting for ascertainment biases.

Methodology
The study utilized a multi-step computational pipeline integrating data from gnomAD v4 and the GenomeIndia project:

Gene Panel Curation: A panel of 696 autosomal-recessive IEM genes was derived from the Genomics England PanelApp (R98), mapped to GRCh38, and reconciled with gnomAD nomenclature.
Data Sources:
- Primary Cohort: gnomAD v4 South Asian (SAS) data comprising 86,254 individuals.
- Validation Cohort: GenomeIndia summary statistics from 9,768 unrelated individuals across 83 endogamous groups.
Variant Classification: Pathogenic variants were categorized into three sets:
- ClinVar Pathogenic/Likely Pathogenic (P/LP) without conflicting interpretations.
- LOFTEE High-Confidence (HC) loss-of-function (LoF) variants.
- Predicted-damaging missense variants (AlphaMissense "likely pathogenic" with REVEL ≥ 0.75).
- Strict quality filters were applied, including a homozygote-only artifact filter and an allele frequency (AF) ceiling of 0.5% to exclude common hypomorphic variants (e.g., MTHFR c.677C>T).
Genetic Modeling:
- Random Mating: Prevalence was calculated using Hardy-Weinberg Equilibrium (HWE), accounting for both true homozygotes and compound heterozygotes ( $GP = q^2$ ).
- Consanguinity Modeling: Recognizing the heterogeneity of inbreeding in South Asia, the study modeled prevalence across a spectrum of Wright's inbreeding coefficients ( $F$ ): 0 (random), 0.01, 0.02 (population average), 0.03, and 0.0625 (first-cousin unions).
- Calibration: Estimates were calibrated against newborn screening (NBS) registries and adjusted for structural variant discovery gaps (CNVs/indels undetectable by short-read sequencing).
Cross-Ancestry Comparison: The pipeline was replicated across eight global gnomAD ancestry groups to assess ascertainment bias and compare carrier frequencies.

Key Results

Carrier Burden: Under random mating, 84.2% of South Asians carry at least one pathogenic allele across the 696 genes. This represents the highest collective carrier frequency among eight global ancestries, closely followed by African (84.0%) and East Asian (83.5%) populations, and significantly higher than founder populations like Finnish (64.8%) and Ashkenazi Jewish (63.6%).
Birth Prevalence (Random Mating): The estimated genetic birth prevalence is 255 per 100,000 live births (1 in 392).
- Breakdown by variant type: 40.6 per 100,000 (ClinVar P/LP), 41.3 per 100,000 (HC LoF), and 62.7 per 100,000 (predicted damaging missense).
- Compound Heterozygosity: Approximately 43% of the total burden (110.5 per 100,000) arises from compound heterozygosity, highlighting the importance of biallelic matrix modeling over simple homozygote counts.
Impact of Consanguinity: Prevalence increases dramatically with inbreeding. Under the modeled population average ( $F=0.02$ ), prevalence rises to 2,072 per 100,000 (1 in 48). For offspring of first-cousin unions ( $F=0.0625$ ), the rate reaches 5,824 per 100,000 (1 in 17).
Top Contributing Genes: No single gene accounts for the majority of the burden. Major contributors include MAN2B2, DPYD, PYGL, ATP7B, PAH, MMUT, PYGM, DHCR24, SPG7, SACS, GLDC, and MLYCD.
Validation (GenomeIndia): In the GenomeIndia cohort, the prevalence for shared variants was 139.6 per 100,000, rising to 151.5 per 100,000 when including GenomeIndia-specific ClinVar P/LP alleles. While lower than the gnomAD SAS estimate (likely due to smaller sample size and reduced detection of ultra-rare alleles), the per-gene allele loads were strongly correlated (Spearman $\rho = 0.83$ ). A pooled estimate yielded 189.1 per 100,000.
Ascertainment Bias: A ClinVar-only analysis significantly underestimated South Asian burden (41 per 100,000) while overestimating Non-Finnish European burden, demonstrating a Eurocentric ascertainment bias. The inclusion of predicted damaging variants and HC LoF variants corrected this imbalance.
Calibration: Applying a clinical-to-genetic scaling factor derived from newborn screening (0.65×) and accounting for structural variant gaps (1.59×) provided exploratory bounds, with a calibrated incidence estimate of 165 per 100,000 and an upper bound of 406 per 100,000.

Significance and Claims
The paper claims that the collective burden of AR-IEMs in South Asia is substantial, with nearly 85% of the population carrying at least one pathogenic allele. The study asserts that while the baseline genetic prevalence under random mating is comparable to other outbred global populations, the clinical prevalence is likely driven significantly by consanguinity and endogamy, which can increase the risk of affected births by an order of magnitude.

Key contributions include:

Correction of Bias: Demonstrating that reliance on ClinVar alone introduces Eurocentric bias, underestimating South Asian risk, and that integrating computational predictors (AlphaMissense, REVEL) and HC LoF filters provides a more equitable and accurate genomic model.
Granular Modeling: Moving beyond a single prevalence estimate to provide a range of scenarios based on inbreeding coefficients ( $F$ ), acknowledging the heterogeneity of South Asian sub-populations.
Clinical Utility: Identifying specific high-burden genes (e.g., DPYD, ATP7B, PAH) that should be prioritized for expanded carrier screening and newborn screening programs tailored to South Asian populations, noting that European-centric panels may have reduced sensitivity.
Reproducibility: Providing a fully reproducible framework, including code, data, and variant-level tables, to facilitate further research and policy development.

The author concludes that these findings provide a critical evidence base for genomic policy, carrier screening strategies, and newborn screening expansion across South Asia, emphasizing the need for population-specific approaches to manage the high burden of recessive metabolic disorders.

More like this