Understanding local population genetics improves variant interpretation in rare genetic disorders

Understanding local population genetics improves variant interpretation in rare genetic disorders | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Brief Communication Understanding local population genetics improves variant interpretation in rare genetic disorders Shahryar Alavi, Zahra Firoozfar, Seyed Mohammad Seyedhassani, and 5 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7448520/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Exome sequencing in underrepresented populations reveals unique variant spectra that improve interpretation of rare genetic disorders. We introduce IRExom, a registry of over 3,000 exomes from Iranian patients enriched for rare diseases, providing population-specific frequencies and constraint metrics. Comparison with gnomAD highlights thousands of variants absent globally but frequent locally, resolving uncertain variants and uncovering founder effects, advancing equitable genomic medicine. Biological sciences/Genetics/Population genetics Biological sciences/Genetics/Genomics Biological sciences/Genetics/Neurodevelopmental disorders Figures Figure 1 Figure 2 Main Exome sequencing is widely used to diagnose rare genetic disorders, but many patients remain unresolved. A major limitation is the underrepresentation of certain populations, including West Asians, in global reference datasets, leading to uncertainty in interpretation of novel variants 1 . We addressed this by establishing IRExom, a population-scale exome resource from 3,392 Iranians collected since January 2019, predominantly from patients with congenital disorders (Fig. 1a). Analysis of probability of LoF intolerance (pLI) in IRExom reveals a distinct and complementary perspective to population-based model of gnomAD 2 (Fig. 1b). While a general correlation exists between the two scoring systems, a notable subset of genes exhibits a significant discrepancy. This difference is not merely a product of our simplified methodology; rather, it highlights a key advantage of the IRExom framework, which is a patient-derived dataset. By relying solely on observed LoF variants, our approach is robust in its ability to assign confidence to genes with LoF variants that may be of unknown consequence. Together with our ACMG-derived variant classification, this provided a practical significance to novel variant detection that was not achievable using the gnomAD constraint model (supplementary Table 1). For instance, we successfully detected two causative variants in the former unknown PPFIBP1 gene—one nonsense and one intronic—with high confidence 3 . Using this approach, we reclassified over 100 variants to Pathogenic/Likely pathogenic or Benign, and submitted these updates to ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/submitters/508498/). To project a comparison between IRExom and global resources, we performed a stringent filtering to keep high-quality variants. This yielded 832,702 SNVs from IRExom, 108,686 (13.1%) of which are not available in none of the global resources dbSNP, gnomAD, or ClinVar (Fig. 1c). Surprisingly, 297,573 (35.7%) of IRExom SNVs are absent from gnomAD, and 91.5% of these are either not reported or classified as uncertain/conflicting in ClinVar 4 (Fig. 1d and Fig. 1e and Extended Data Fig. 1a) of which 15,356 have frequencies greater than 0.01 in IRExom (Extended Data Fig. 1b and Extended Data Fig. 1c). Indeed, this might be due to IRExom phenotypic composition, thus we inspected the number of homozygote individuals for all IRExom variants that are not reported or classified as uncertain/conflicting in ClinVar, with their frequencies in gnomAD (Fig. 1f). We observed that 16,544 IRExom variants are homozygous in more than 5 IRExom individuals and their gnomAD frequency is less than 1e-5 (or absent from gnomAD) without a decisive ClinVar record. However, according to ACMG/AMP 2015 guidelines 5 , such variants could be misclassified without local reference data, demonstrating that absence from gnomAD does not imply pathogenicity. Beyond short variants, IRExom supports detection of copy number variations (CNVs) by creating a reference panel. Our CNV detection tool, PalinDepth, identified homozygous deletions in genes confirmed clinically. Interestingly, our we found a frequent exon deletion in the SGCB gene in IRExom eastern subpopulation, which has been previously confirmed as a founder effect using other methods 6 . These findings can refine genetic counselling by anticipating carrier burden and recurrence risk. Finally, we assessed population structure within our West Asian cohort, known as highly consanguineous. To this purpose, we further filtered IRExom samples, remaining 1,389 individuals. Population stratification showed distinct clusters corresponding to Iranian ethnic groups, as well as an admixed group at their intersection (Fig. 2a). This is consistent with previous findings, showing genetically distinct neighbours despite high levels of inbreeding in the population, highlighting the enriched diverse ancestry that has shaped the Iranian population 7 . In addition, analysis of runs of homozygosity (ROH) showed eastern subpopulation has long ROH segments even at low ROH segment counts, consistent with endogamy, while IRExom western subpopulation displayed short, fewer ROH segments, reflecting greater outbreeding (Fig. 2b). These findings underscore both the diversity in degrees of consanguinity among Iranians and their relevance for recessive disease risk assessment. Together, IRExom demonstrates that regional exome aggregation provides a versatile genomic resource and substantially improves both population genetics and rare disease diagnostics. Indeed one of our limitations is that IRExom is built on targeted short read sequence data, and we believe that introducing long read sequencing into underrepsented populations, as shown recently, would further sheds light on hidden aspects of human genome 8 . By providing constraint metrics, allele frequencies, and population structure insights, it informs ACMG-guided classification, enables discovery of founder mutations, and advises local genetic counselling. IRExom thus establishes a scalable model for integrating underrepresented populations into global efforts in variant curation, gene discovery, and population genetics. Methods DNA extracted from peripheral blood to be sequenced on Illumina NovaSeq platform using Agilent SureSelect Human All Exon V7 capture kit in Macrogen Inc. South Korea. Exome data were collected by Palindrome from the collaborating medical genetics laboratories. Informed consent from participants and ethics approval from local authorities were obtained by laboratories. All data were anonymised prior to our exome analysis. Joint genotyping Exome FASTQ files were taken for quality control using FastQC Version 0.11.9 9 . Reads were aligned to human genome assembly hg38 using BWA-MEM 10 and germline variants called according to GATK v4 Best Practice and GVCF files were produced 11 . Joint genotyping was performed with GATK GenomicsDB followed by hard filtration of very low-quality variants. The cohort-level dataset was further processed for data sanity at both sample and variant levels. Samples sequenced by dispersed exome capture kits were removed (keeping only ones sequenced by the above mentioned SureSelect V7 kit). Then, multiallelic and INDEL variant loci were excluded and only variants with QUAL value greater than 500 were kept. The resulting biallelic SNVs VCF file was employed for annotation using global genomic datasets dbSNP build 155, gnomAD v4.1 Exomes, and ClinVar 20250810. Constraint calculation Annotated VCF was used to retrieve a list of LoF variants of each gene, and counting the number of pathogenic, benign, and total variants. We developed an R script to quantify the degree of intolerance of each gene to LoF variants based on the ratio of pathogenic to benign IRExom LoF variants. An LoF variant was defined as a variant with one of these consequences: splicing, frameshift, start loss, or stop gain. The clinical significance was extracted from their ClinVar annotation, then the IRExom pLI is calculated for each gene using this formula: IRExom pLI = 1/2 * (1 – (N Benign − N Pathogenic )/N Total ) Where: N Benign is the count of Benign and Likely benign LoF variants in the gene, N Pathogenic is the count of Pathogenic and Likely pathogenic LoF variants in the gene, and N Total is the sum of all LoF variants in the gene (regardless of their ClinVar status). The score is normalised to a value between 0 and 1, provides a robust, clinically-informed metric for assessing gene constraint based on existing variant annotations. A score of 0.5 indicates a balanced ratio between benign and pathogenic variants, suggesting a gene that has an unknown LoF significance. A score approaching 1 signifies that the gene is highly intolerant to LoF. Conversely, a score approaching 0 indicates that the gene's LOF variants are primarily benign. Genes with no ClinVar classified LoF variants (or ClinVar classified as uncertain) were assigned an intolerance score of 0.5, too. Population stratification The cohort-level dataset was further processed by piping into PLINK 2.0 12 with these arguments: --geno 0.1 --mind 0.1 --snps-only just-acgt. Using KING 13 individuals with kinships to 3 rd degree were found and excluded. Samples with heterozygosity rate ±3SD (standard deviation) and with outlier genotype missing rate were removed. Linkage disequilibrium was calculated using --indep-pairwise 50 5 0.2 and variant were pruned. The output was used for principal component analysis (PCA; --pca 40), and homozygosity mapping with these arguments: --homozyg-snp 50 --homozyg-kb 500 --homozyg-window-snp 30. We developed a Python tool to utilise the PCA result by uniform manifold approximation and projection (UMAP) with hierarchical density-based spatial clustering of applications with noise (HDBSCAN) to perform a grid search for the best clustering of IRExom population. Our UMAP grid search tool gets a list of values for each of these parameters: number of PCs, number of neighbours, minimum distance, minimum samples, minimum cluster size. By making all possible combinations, the tool calculates silhouette score to find the best clustering. Declarations Data availability The conceptual analysis scripts are available online at IRExom GitHub repository (https://github.com/Schahrjar/IRExom) for the community. Genotype level data is not available due to ethics limitations. References Corpas, M., Guio, H., Lopez-Correa, C. & Fatumo, S. Why genomic diversity should not be framed by census alone. Nature Genetics 2025 57:8 57 , 1793–1794 (2025). Chen, S. et al. A genomic mutational constraint map using variation in 76,156 human genomes. Nature 2023 625:7993 625 , 92–100 (2023). Rosenhahn, E. et al. Bi-allelic loss-of-function variants in PPFIBP1 cause a neurodevelopmental disorder with microcephaly, epilepsy, and periventricular calcifications. Am J Hum Genet 109 , 1421–1435 (2022). Landrum, M. J. et al. ClinVar: public archive of relationships among sequence variation and human phenotype. Nucleic Acids Res 42 , D980–D985 (2014). Richards, S. et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genetics in Medicine 2015 17:5 17 , 405–423 (2015). Mojbafan, M., Bahmani, R., Bagheri, S. D., Sharifi, Z. & Zeinali, S. Mutational spectrum of autosomal recessive limb-girdle muscular dystrophies in a cohort of 112 Iranian patients and reporting of a possible founder effect. Orphanet J Rare Dis 15 , 1–10 (2020). Mehrjoo, Z. et al. Distinct genetic variation and heterogeneity of the Iranian population. PLoS Genet 15 , e1008385 (2019). Ghorbani, M. et al. Near-complete Middle Eastern genomes refine autozygosity and enhance disease-causing and population-specific variant discovery. Nature Genetics 2025 57:5 57 , 1119–1131 (2025). Babraham Bioinformatics - FastQC A Quality Control tool for High Throughput Sequence Data. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/. Li, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. (2013). Poplin, R. et al. Scaling accurate genetic variant discovery to tens of thousands of samples. bioRxiv 201178 (2018) doi:10.1101/201178. Chang, C. C. et al. Second-generation PLINK: Rising to the challenge of larger and richer datasets. Gigascience 4 , 7 (2015). Manichaikul, A. et al. Robust relationship inference in genome-wide association studies. Bioinformatics 26 , 2867–2873 (2010). Additional Declarations There is NO Competing Interest. Supplementary Files IRExomSupplementaryTable.xlsx IRExom pLI scores ExtendedDataFig.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7448520","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Brief Communication","associatedPublications":[],"authors":[{"id":507096584,"identity":"448c6ff3-0835-4d28-8c6f-ab7f8b6a8f8e","order_by":0,"name":"Shahryar Alavi","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAxklEQVRIiWNgGAWjYHACxgMMDBJyDAzMDWCuBDF6QFqMgVpJ08KQ2EC0FvkG5gcHfrZZpG84frCB4UcNQ+LMBgJaDA6wGRzsbZPI3XAmsYGx5xhD4mxCthiAdPGcAWq5AXQYbwND4jzCDmP/cPDPGYl0A6AWxr/EaGE4wGNwmKdCIgGkhRlkC2GHHeYpOCxTIWE4E+iXwzLHJIwJel++vX3jwzcGdfJ8xw8ffPimxkZ2xgFC1jAjO5LIiBwFo2AUjIJRQAgAALBbQCyWEzOuAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-3484-3423","institution":"UCL Queen Square Institute of Neurology","correspondingAuthor":true,"prefix":"","firstName":"Shahryar","middleName":"","lastName":"Alavi","suffix":""},{"id":507096585,"identity":"31a52bd5-baa4-4a2f-9235-3fe77c06561e","order_by":1,"name":"Zahra Firoozfar","email":"","orcid":"https://orcid.org/0009-0009-3927-6796","institution":"Palindrome","correspondingAuthor":false,"prefix":"","firstName":"Zahra","middleName":"","lastName":"Firoozfar","suffix":""},{"id":507096586,"identity":"e1073306-6fdb-47c9-9efd-267fe963ef9c","order_by":2,"name":"Seyed Mohammad Seyedhassani","email":"","orcid":"","institution":"Dr. Seyedhassani Medical Genetic Center","correspondingAuthor":false,"prefix":"","firstName":"Seyed","middleName":"Mohammad","lastName":"Seyedhassani","suffix":""},{"id":507096587,"identity":"0c16c576-325d-4f9f-bab9-0eee586386e9","order_by":3,"name":"Pegah Moshtaghian","email":"","orcid":"https://orcid.org/0009-0005-5700-8542","institution":"Palindrome","correspondingAuthor":false,"prefix":"","firstName":"Pegah","middleName":"","lastName":"Moshtaghian","suffix":""},{"id":507096588,"identity":"72fa3a1f-cf32-4c89-be9b-803a63e1117c","order_by":4,"name":"Mohammad Yahya Vahidi Mehrjardi","email":"","orcid":"","institution":"Avin Medical Genetics Centre","correspondingAuthor":false,"prefix":"","firstName":"Mohammad","middleName":"Yahya Vahidi","lastName":"Mehrjardi","suffix":""},{"id":507096589,"identity":"4fbe9f9c-bdda-40c3-9b9a-5ddc4b97f193","order_by":5,"name":"Mostafa Montazer Zohour","email":"","orcid":"","institution":"Mendel Medical Genetics Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Mostafa","middleName":"Montazer","lastName":"Zohour","suffix":""},{"id":507096590,"identity":"04c24190-2696-4725-b275-157d8d866cfc","order_by":6,"name":"Reza Alibakhshi","email":"","orcid":"","institution":"Dr Alibakhshi Medical Genetics Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Reza","middleName":"","lastName":"Alibakhshi","suffix":""},{"id":507096591,"identity":"d5ac583a-8640-4f42-8b32-5e4a8aa29015","order_by":7,"name":"Mehryar Alavi","email":"","orcid":"","institution":"Palindrome","correspondingAuthor":false,"prefix":"","firstName":"Mehryar","middleName":"","lastName":"Alavi","suffix":""}],"badges":[],"createdAt":"2025-08-24 23:40:11","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7448520/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7448520/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":92495980,"identity":"e76f7cc0-cc63-4618-ac34-b1d53dcf0572","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":2924506,"visible":true,"origin":"","legend":"","description":"","filename":"IRExommanuscriptNautreGenetics.docx","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/1a62703758fcea1fc2ed6f10.docx"},{"id":92496864,"identity":"f789d4df-b008-44c6-b314-57fdb8cd9819","added_by":"auto","created_at":"2025-09-30 10:33:38","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":342,"visible":true,"origin":"","legend":"","description":"","filename":"rs7448520.json","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/d196c8ebff9d3ab9ce795026.json"},{"id":92495968,"identity":"d2a32ce6-873c-4095-b4e4-9c16423b7733","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"xml","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":35684,"visible":true,"origin":"","legend":"","description":"","filename":"rs74485202enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/88f577ca858981b9c08d9ba4.xml"},{"id":92497800,"identity":"883bde4b-568e-4b4d-83d1-ff7ae38522ed","added_by":"auto","created_at":"2025-09-30 10:41:38","extension":"jpeg","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1021354,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/e181fb159533114888c45df1.jpeg"},{"id":92495979,"identity":"7960110c-558f-4bcd-9296-e2b25a2d91bf","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":617583,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/fd3cfb4ab1a61db77642d88e.jpeg"},{"id":92495977,"identity":"8e6c0136-ba0c-4b7f-8c84-7d33a4770444","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1154786,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/d336949371c2d89924282483.jpeg"},{"id":92495973,"identity":"f3411a33-a1d6-414f-aee1-c3f2f85208ad","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"png","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":127602,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/1b07eeeaa2d7c82d8f2f52c4.png"},{"id":92496866,"identity":"8dea01a9-c109-412a-944c-b26fcb199a7d","added_by":"auto","created_at":"2025-09-30 10:33:38","extension":"png","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":78032,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/a6734d427dbf072136c4c4a8.png"},{"id":92498115,"identity":"d869187f-eb6d-4f73-bf5a-baf7fa141405","added_by":"auto","created_at":"2025-09-30 10:49:38","extension":"png","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":135313,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/4999f716555c887f222f4def.png"},{"id":92496867,"identity":"f63c4a7c-2049-40c9-9d1a-5461b1898b47","added_by":"auto","created_at":"2025-09-30 10:33:38","extension":"xml","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":34169,"visible":true,"origin":"","legend":"","description":"","filename":"rs74485202structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/46427062dc30cec7e1c875a8.xml"},{"id":92496869,"identity":"76b8ce16-509e-493b-a215-169d66114445","added_by":"auto","created_at":"2025-09-30 10:33:38","extension":"html","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":41351,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/44655f996a7a790c3682e714.html"},{"id":92495966,"identity":"02647629-c628-4a10-959b-fcfa389f0f65","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":190507,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIRExom statistics and its variants in global resources.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, summary of 9 major phenotypic categories that are recruited in IRExom. Darker bars (left bars in each category) are count of males and light bars (right bars) are count of females. Negative diagnosis assigned to cases where there is not any detected variant, even VUSes, compatible with the patient’s phenotype. \u003cstrong\u003eb\u003c/strong\u003e, comparison of IRExom pLI with gnomAD LOEUF. To be comparable in a range of 0-1, reversed percentile of gnomAD LOEUF was calculated. IRExom pLI is a continuous metric of genes' constraint, with higher values indicating a greater proportion of pathogenic variants over all LoF variants of that gene. Genes with unknown LoF significance have IRExom pLI of 0.5. \u003cstrong\u003ec\u003c/strong\u003e, statistics of IRExom exclusive SNVs along with global databases dbSNP, gnomAD, and ClinVar. \u003cstrong\u003ed\u003c/strong\u003e, IRExom high fidelity SNVs’ frequencies against gnomAD frequencies, coloured by their significance reported to ClinVar. For a better visualisation, variants not reported to ClinVar are removed from this plot (see Extended Data Fig. 1a which includes variants not reported to ClinVar). The X-axis is artificially extended beyond 10e-6 (the lowest gnomAD frequency) to visualise IRExom variants that are absent from gnomAD (the strip of variants at the left side of the scatter plot). \u003cstrong\u003ee\u003c/strong\u003e, number of IRExom SNVs absent from gnomAD, categorised by their ClinVar significance. \u003cstrong\u003ef\u003c/strong\u003e, number of IRExom homozygote individuals for SNVs not reported or uncertain/conflicting in ClinVar, with their frequencies in gnomAD.\u003c/p\u003e","description":"","filename":"1.png","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/c549ed3dae41ada058d49359.png"},{"id":92495965,"identity":"7615f807-0c07-4c5b-b336-536c48a8ab8f","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":96496,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eIRExom population structure.\u003c/strong\u003e \u003cstrong\u003ea\u003c/strong\u003e, best UMAP of IRExom population with 5 major genetic contexts detected (silhouette score = 0.6). We named each cluster by the geographic directions that we received the most of cluster’s sample data from, since no definitive ancestry has been claimed in IRExom registry. UMAP mainly looks for similarities to find neighbours, and distance between clusters is not an indicative of isolation. \u003cstrong\u003eb\u003c/strong\u003e, ROH burden analysis across Iranian subpopulations. Percentages in the quartiles show the proportion of each subpopulation fallen into that quartile.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/9f2c17b84c372851f4a56fc5.png"},{"id":92498116,"identity":"168b827d-0f89-491b-898b-83f72ef3e6b5","added_by":"auto","created_at":"2025-09-30 10:49:43","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":636752,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/60f3a4a5-1a66-4802-ade0-364e1a5bc1e4.pdf"},{"id":92495970,"identity":"1b7d5a58-7b9b-4c94-b41a-82d02eded8e0","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"xlsx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":353184,"visible":true,"origin":"","legend":"IRExom pLI scores","description":"","filename":"IRExomSupplementaryTable.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/c33129bf28076be382b68f60.xlsx"},{"id":92495972,"identity":"1d5eb709-964e-4b71-8adc-c9b5550e4080","added_by":"auto","created_at":"2025-09-30 10:25:38","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":1169103,"visible":true,"origin":"","legend":"","description":"","filename":"ExtendedDataFig.docx","url":"https://assets-eu.researchsquare.com/files/rs-7448520/v1/d64f3fe453ba44c0a1f39bb4.docx"}],"financialInterests":"There is \u003cb\u003eNO\u003c/b\u003e Competing Interest.","formattedTitle":"Understanding local population genetics improves variant interpretation in rare genetic disorders","fulltext":[{"header":"Main","content":"\u003cp\u003eExome sequencing is widely used to diagnose rare genetic disorders, but many patients remain unresolved. A major limitation is the underrepresentation of certain populations, including West Asians, in global reference datasets, leading to uncertainty in interpretation of novel variants\u003csup\u003e1\u003c/sup\u003e. We addressed this by establishing IRExom, a population-scale exome resource from 3,392 Iranians collected since January 2019, predominantly from patients with congenital disorders (Fig. 1a).\u003c/p\u003e\n\u003cp\u003eAnalysis of probability of LoF intolerance (pLI) in IRExom reveals a distinct and complementary perspective to population-based model of gnomAD\u003csup\u003e2\u003c/sup\u003e (Fig. 1b). While a general correlation exists between the two scoring systems, a notable subset of genes exhibits a significant discrepancy. This difference is not merely a product of our simplified methodology; rather, it highlights a key advantage of the IRExom framework, which is a patient-derived dataset. By relying solely on observed LoF variants, our approach is robust in its ability to assign confidence to genes with LoF variants that may be of unknown consequence. Together with our ACMG-derived variant classification, this provided a practical significance to novel variant detection that was not achievable using the gnomAD constraint model (supplementary Table 1). For instance, we successfully detected two causative variants in the former unknown PPFIBP1 gene—one nonsense and one intronic—with high confidence\u003csup\u003e3\u003c/sup\u003e. Using this approach, we reclassified over 100 variants to Pathogenic/Likely pathogenic or Benign, and submitted these updates to ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/submitters/508498/).\u003c/p\u003e\n\u003cp\u003eTo project a comparison between IRExom and global resources, we performed a stringent filtering to keep high-quality variants. This yielded 832,702 SNVs from IRExom, 108,686 (13.1%) of which are not available in none of the global resources dbSNP, gnomAD, or ClinVar (Fig. 1c). Surprisingly, 297,573 (35.7%) of IRExom SNVs are absent from gnomAD, and 91.5% of these are either not reported or classified as uncertain/conflicting in ClinVar\u003csup\u003e4\u003c/sup\u003e (Fig. 1d and Fig. 1e and Extended Data Fig. 1a) of which 15,356 have frequencies greater than 0.01 in IRExom (Extended Data Fig. 1b and Extended Data Fig. 1c). Indeed, this might be due to IRExom phenotypic composition, thus we inspected the number of homozygote individuals for all IRExom variants that are not reported or classified as uncertain/conflicting in ClinVar, with their frequencies in gnomAD (Fig. 1f). We observed that 16,544 IRExom variants are homozygous in more than 5 IRExom individuals and their gnomAD frequency is less than 1e-5 (or absent from gnomAD) without a decisive ClinVar record. However, according to ACMG/AMP 2015 guidelines\u003csup\u003e5\u003c/sup\u003e, such variants could be misclassified without local reference data, demonstrating that absence from gnomAD does not imply pathogenicity.\u003c/p\u003e\n\u003cp\u003eBeyond short variants, IRExom supports detection of copy number variations (CNVs) by creating a reference panel. Our CNV detection tool, PalinDepth, identified homozygous deletions in genes confirmed clinically. Interestingly, our we found a frequent exon deletion in the \u003cem\u003eSGCB\u003c/em\u003e gene in IRExom eastern subpopulation, which has been previously confirmed as a founder effect using other methods\u003csup\u003e6\u003c/sup\u003e. These findings can refine genetic counselling by anticipating carrier burden and recurrence risk.\u003c/p\u003e\n\u003cp\u003eFinally, we assessed population structure within our West Asian cohort, known as highly consanguineous. To this purpose, we further filtered IRExom samples, remaining 1,389 individuals. Population stratification showed distinct clusters corresponding to Iranian ethnic groups, as well as an admixed group at their intersection (Fig. 2a). This is consistent with previous findings, showing genetically distinct neighbours despite high levels of inbreeding in the population, highlighting the enriched diverse ancestry that has shaped the Iranian population\u003csup\u003e7\u003c/sup\u003e. In addition, analysis of runs of homozygosity (ROH) showed eastern subpopulation has long ROH segments even at low ROH segment counts, consistent with endogamy, while IRExom western subpopulation displayed short, fewer ROH segments, reflecting greater outbreeding (Fig. 2b). These findings underscore both the diversity in degrees of consanguinity among Iranians and their relevance for recessive disease risk assessment.\u003c/p\u003e\n\u003cp\u003eTogether, IRExom demonstrates that regional exome aggregation provides a versatile genomic resource and substantially improves both population genetics and rare disease diagnostics. Indeed one of our limitations is that IRExom is built on targeted short read sequence data, and we believe that introducing long read sequencing into underrepsented populations, as shown recently, would further sheds light on hidden aspects of human genome\u003csup\u003e8\u003c/sup\u003e. By providing constraint metrics, allele frequencies, and population structure insights, it informs ACMG-guided classification, enables discovery of founder mutations, and advises local genetic counselling. IRExom thus establishes a scalable model for integrating underrepresented populations into global efforts in variant curation, gene discovery, and population genetics.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eDNA extracted from peripheral blood to be sequenced on Illumina NovaSeq platform using Agilent SureSelect Human All Exon V7 capture kit in Macrogen Inc. South Korea. Exome data were collected by Palindrome from the collaborating medical genetics laboratories. Informed consent from participants and ethics approval from local authorities were obtained by laboratories. All data were anonymised prior to our exome analysis.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eJoint genotyping\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExome FASTQ files were taken for quality control using FastQC Version 0.11.9\u003csup\u003e9\u003c/sup\u003e. Reads were aligned to human genome assembly hg38 using BWA-MEM\u003csup\u003e10\u003c/sup\u003e and germline variants called according to GATK v4 Best Practice and GVCF files were produced\u003csup\u003e11\u003c/sup\u003e. Joint genotyping was performed with GATK GenomicsDB followed by hard filtration of very low-quality variants. The cohort-level dataset was further processed for data sanity at both sample and variant levels.\u003c/p\u003e\n\u003cp\u003eSamples sequenced by dispersed exome capture kits were removed (keeping only ones sequenced by the above mentioned SureSelect V7 kit). Then, multiallelic and INDEL variant loci were excluded and only variants with QUAL value greater than 500 were kept. The resulting biallelic SNVs VCF file was employed for annotation using global genomic datasets dbSNP build 155, gnomAD v4.1 Exomes, and ClinVar 20250810.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConstraint calculation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAnnotated VCF was used to retrieve a list of LoF variants of each gene, and counting the number of pathogenic, benign, and total variants. We developed an R script to quantify the degree of intolerance of each gene to LoF variants based on the ratio of pathogenic to benign IRExom LoF variants.\u003c/p\u003e\n\u003cp\u003eAn LoF variant was defined as a variant with one of these consequences: splicing, frameshift, start loss, or stop gain. The clinical significance was extracted from their ClinVar annotation, then the IRExom pLI is calculated for each gene using this formula:\u003c/p\u003e\n\u003cp\u003eIRExom pLI = 1/2 * (1 \u0026ndash; (N\u003csub\u003eBenign\u003c/sub\u003e \u0026minus; N\u003csub\u003ePathogenic\u003c/sub\u003e)/N\u003csub\u003eTotal\u003c/sub\u003e)\u003c/p\u003e\n\u003cp\u003eWhere: N\u003csub\u003eBenign\u003c/sub\u003e is the count of Benign and Likely benign LoF variants in the gene, N\u003csub\u003ePathogenic\u003c/sub\u003e is the count of Pathogenic and Likely pathogenic LoF variants in the gene, and N\u003csub\u003eTotal\u003c/sub\u003e is the sum of all LoF variants in the gene (regardless of their ClinVar status).\u003c/p\u003e\n\u003cp\u003eThe score is normalised to a value between 0 and 1, provides a robust, clinically-informed metric for assessing gene constraint based on existing variant annotations. A score of 0.5 indicates a balanced ratio between benign and pathogenic variants, suggesting a gene that has an unknown LoF significance. A score approaching 1 signifies that the gene is highly intolerant to LoF. Conversely, a score approaching 0 indicates that the gene\u0026apos;s LOF variants are primarily benign. Genes with no ClinVar classified LoF variants (or ClinVar classified as uncertain) were assigned an intolerance score of 0.5, too.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003ePopulation stratification\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe cohort-level dataset was further processed by piping into PLINK 2.0\u003csup\u003e12\u003c/sup\u003e with these arguments: --geno 0.1 --mind 0.1 --snps-only just-acgt. Using KING\u003csup\u003e13\u003c/sup\u003e individuals with kinships to 3\u003csup\u003erd\u003c/sup\u003e degree were found and excluded. Samples with heterozygosity rate \u0026plusmn;3SD (standard deviation) and with outlier genotype missing rate were removed. Linkage disequilibrium was calculated using --indep-pairwise 50 5 0.2 and variant were pruned. The output was used for principal component analysis (PCA; --pca 40), and homozygosity mapping with these arguments: --homozyg-snp 50 --homozyg-kb 500 --homozyg-window-snp 30.\u003c/p\u003e\n\u003cp\u003eWe developed a Python tool to utilise the PCA result by uniform manifold approximation and projection (UMAP) with hierarchical density-based spatial clustering of applications with noise (HDBSCAN) to perform a grid search for the best clustering of IRExom population. Our UMAP grid search tool gets a list of values for each of these parameters: number of PCs, number of neighbours, minimum distance, minimum samples, minimum cluster size. By making all possible combinations, the tool calculates silhouette score to find the best clustering.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe conceptual analysis scripts are available online at IRExom GitHub repository (https://github.com/Schahrjar/IRExom) for the community.\u003c/p\u003e\n\u003cp\u003eGenotype level data is not available due to ethics limitations.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eCorpas, M., Guio, H., Lopez-Correa, C. \u0026amp; Fatumo, S. Why genomic diversity should not be framed by census alone. \u003cem\u003eNature Genetics 2025 57:8\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 1793\u0026ndash;1794 (2025).\u003c/li\u003e\n\u003cli\u003eChen, S. \u003cem\u003eet al.\u003c/em\u003e A genomic mutational constraint map using variation in 76,156 human genomes. \u003cem\u003eNature 2023 625:7993\u003c/em\u003e \u003cstrong\u003e625\u003c/strong\u003e, 92\u0026ndash;100 (2023).\u003c/li\u003e\n\u003cli\u003eRosenhahn, E. \u003cem\u003eet al.\u003c/em\u003e Bi-allelic loss-of-function variants in PPFIBP1 cause a neurodevelopmental disorder with microcephaly, epilepsy, and periventricular calcifications. \u003cem\u003eAm J Hum Genet\u003c/em\u003e \u003cstrong\u003e109\u003c/strong\u003e, 1421\u0026ndash;1435 (2022).\u003c/li\u003e\n\u003cli\u003eLandrum, M. J. \u003cem\u003eet al.\u003c/em\u003e ClinVar: public archive of relationships among sequence variation and human phenotype. \u003cem\u003eNucleic Acids Res\u003c/em\u003e \u003cstrong\u003e42\u003c/strong\u003e, D980\u0026ndash;D985 (2014).\u003c/li\u003e\n\u003cli\u003eRichards, S. \u003cem\u003eet al.\u003c/em\u003e Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. \u003cem\u003eGenetics in Medicine 2015 17:5\u003c/em\u003e \u003cstrong\u003e17\u003c/strong\u003e, 405\u0026ndash;423 (2015).\u003c/li\u003e\n\u003cli\u003eMojbafan, M., Bahmani, R., Bagheri, S. D., Sharifi, Z. \u0026amp; Zeinali, S. Mutational spectrum of autosomal recessive limb-girdle muscular dystrophies in a cohort of 112 Iranian patients and reporting of a possible founder effect. \u003cem\u003eOrphanet J Rare Dis\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, 1\u0026ndash;10 (2020).\u003c/li\u003e\n\u003cli\u003eMehrjoo, Z. \u003cem\u003eet al.\u003c/em\u003e Distinct genetic variation and heterogeneity of the Iranian population. \u003cem\u003ePLoS Genet\u003c/em\u003e \u003cstrong\u003e15\u003c/strong\u003e, e1008385 (2019).\u003c/li\u003e\n\u003cli\u003eGhorbani, M. \u003cem\u003eet al.\u003c/em\u003e Near-complete Middle Eastern genomes refine autozygosity and enhance disease-causing and population-specific variant discovery. \u003cem\u003eNature Genetics 2025 57:5\u003c/em\u003e \u003cstrong\u003e57\u003c/strong\u003e, 1119\u0026ndash;1131 (2025).\u003c/li\u003e\n\u003cli\u003eBabraham Bioinformatics - FastQC A Quality Control tool for High Throughput Sequence Data. https://www.bioinformatics.babraham.ac.uk/projects/fastqc/.\u003c/li\u003e\n\u003cli\u003eLi, H. Aligning sequence reads, clone sequences and assembly contigs with BWA-MEM. (2013).\u003c/li\u003e\n\u003cli\u003ePoplin, R. \u003cem\u003eet al.\u003c/em\u003e Scaling accurate genetic variant discovery to tens of thousands of samples. \u003cem\u003ebioRxiv\u003c/em\u003e 201178 (2018) doi:10.1101/201178.\u003c/li\u003e\n\u003cli\u003eChang, C. C. \u003cem\u003eet al.\u003c/em\u003e Second-generation PLINK: Rising to the challenge of larger and richer datasets. \u003cem\u003eGigascience\u003c/em\u003e \u003cstrong\u003e4\u003c/strong\u003e, 7 (2015).\u003c/li\u003e\n\u003cli\u003eManichaikul, A. \u003cem\u003eet al.\u003c/em\u003e Robust relationship inference in genome-wide association studies. \u003cem\u003eBioinformatics\u003c/em\u003e \u003cstrong\u003e26\u003c/strong\u003e, 2867\u0026ndash;2873 (2010).\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7448520/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7448520/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eExome sequencing in underrepresented populations reveals unique variant spectra that improve interpretation of rare genetic disorders. We introduce IRExom, a registry of over 3,000 exomes from Iranian patients enriched for rare diseases, providing population-specific frequencies and constraint metrics. Comparison with gnomAD highlights thousands of variants absent globally but frequent locally, resolving uncertain variants and uncovering founder effects, advancing equitable genomic medicine.\u003c/p\u003e","manuscriptTitle":"Understanding local population genetics improves variant interpretation in rare genetic disorders","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-09-30 10:25:33","doi":"10.21203/rs.3.rs-7448520/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"1ead5758-27a6-4740-a3aa-e2e35adad4ad","owner":[],"postedDate":"September 30th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":53866501,"name":"Biological sciences/Genetics/Population genetics"},{"id":53866502,"name":"Biological sciences/Genetics/Genomics"},{"id":53866503,"name":"Biological sciences/Genetics/Neurodevelopmental disorders"}],"tags":[],"updatedAt":"2025-09-30T10:25:33+00:00","versionOfRecord":[],"versionCreatedAt":"2025-09-30 10:25:33","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7448520","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7448520","identity":"rs-7448520","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}