Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches

Abstract A Type 1 Diabetes Genetic Risk Score (T1DGRS) aids diagnosis and prediction of Type 1 Diabetes (T1D). While traditionally derived from imputed array genotypes, Whole Genome Sequencing (WGS) provides a more direct approach and is now increasingly used in clinical and research studies. We inv...

Full description

Saved in:
Bibliographic Details
Main Authors: Ankit M. Arni, Diane P. Fraser, Seth A. Sharp, Richard A. Oram, Matthew B. Johnson, Michael N. Weedon, Kashyap A. Patel
Format: Article
Language:English
Published: Nature Portfolio 2024-12-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-024-82278-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846101189233475584
author Ankit M. Arni
Diane P. Fraser
Seth A. Sharp
Richard A. Oram
Matthew B. Johnson
Michael N. Weedon
Kashyap A. Patel
author_facet Ankit M. Arni
Diane P. Fraser
Seth A. Sharp
Richard A. Oram
Matthew B. Johnson
Michael N. Weedon
Kashyap A. Patel
author_sort Ankit M. Arni
collection DOAJ
description Abstract A Type 1 Diabetes Genetic Risk Score (T1DGRS) aids diagnosis and prediction of Type 1 Diabetes (T1D). While traditionally derived from imputed array genotypes, Whole Genome Sequencing (WGS) provides a more direct approach and is now increasingly used in clinical and research studies. We investigated the concordance between WGS-based and array-based T1DGRS across genetic ancestries in 149,265 UK Biobank participants using WGS, TOPMed-imputed, and 1000 Genomes-imputed array genotypes. In the overall cohort, WGS-based T1DGRS demonstrated strong correlation with TOPMed-imputed array-based score (r = 0.996, average WGS-based score 0.0028 standard deviations (SD) lower, p < 10− 31), while showing lower correlation with 1000 Genomes-imputed array-based scores (r = 0.981, 0.043 SD lower in WGS, p < 10− 300). Ancestry-stratified analyses between WGS-based and TOPMed-imputed array-based score showed the highest correlation with European ancestry (r = 0.996, 0.044 SD lower in WGS, p < 10− 300) followed by African ancestry (r = 0.989, 0.0193 SD lower in WGS, p < 10− 14) and South Asian ancestry (r = 0.986, 0.0129 SD lower in WGS, p < 10 − 6). These differences were more pronounced when comparing WGS based score with 1000 Genomes-imputed array-based scores (r = 0.982, 0.975, 0.957 for European, South Asian, African respectively). Population-level analysis using WGS-based T1DGRS revealed significant ancestry-based stratification, with European ancestry individuals showing the highest scores, followed by South Asian (average 0.28 SD lower than Europeans, p < 10− 58) and African ancestry individuals (average 0.89 SD lower than Europeans, p < 10− 300). Notably, when applying the European ancestry-derived 90th centile risk threshold, only 0.71% (95% CI 0.41–1.13) of African ancestry individuals and 6.4% (95% CI 5.6–7.2) of South Asian individuals were identified as high-risk, substantially below the expected 10%. In conclusion, while WGS is viable for generating T1DGRS, with TOPMed-imputed genotypes offering a cost-effective alternative, the persistence of ancestry-based variations in T1DGRS distribution even using whole genome sequencing emphasises the need for ancestry-specific or pan-ancestry standards in clinical practice.
format Article
id doaj-art-cbc6294777ee44d5ad0efa40e0fc8ea7
institution Kabale University
issn 2045-2322
language English
publishDate 2024-12-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-cbc6294777ee44d5ad0efa40e0fc8ea72024-12-29T12:27:44ZengNature PortfolioScientific Reports2045-23222024-12-0114111010.1038/s41598-024-82278-xType 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approachesAnkit M. Arni0Diane P. Fraser1Seth A. Sharp2Richard A. Oram3Matthew B. Johnson4Michael N. Weedon5Kashyap A. Patel6Department of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterDepartment of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterDepartment of Pediatrics, Stanford UniversityDepartment of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterDepartment of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterDepartment of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterDepartment of Clinical and Biomedical Sciences, RILD Building, Royal Devon and Exeter Hospital, University of ExeterAbstract A Type 1 Diabetes Genetic Risk Score (T1DGRS) aids diagnosis and prediction of Type 1 Diabetes (T1D). While traditionally derived from imputed array genotypes, Whole Genome Sequencing (WGS) provides a more direct approach and is now increasingly used in clinical and research studies. We investigated the concordance between WGS-based and array-based T1DGRS across genetic ancestries in 149,265 UK Biobank participants using WGS, TOPMed-imputed, and 1000 Genomes-imputed array genotypes. In the overall cohort, WGS-based T1DGRS demonstrated strong correlation with TOPMed-imputed array-based score (r = 0.996, average WGS-based score 0.0028 standard deviations (SD) lower, p < 10− 31), while showing lower correlation with 1000 Genomes-imputed array-based scores (r = 0.981, 0.043 SD lower in WGS, p < 10− 300). Ancestry-stratified analyses between WGS-based and TOPMed-imputed array-based score showed the highest correlation with European ancestry (r = 0.996, 0.044 SD lower in WGS, p < 10− 300) followed by African ancestry (r = 0.989, 0.0193 SD lower in WGS, p < 10− 14) and South Asian ancestry (r = 0.986, 0.0129 SD lower in WGS, p < 10 − 6). These differences were more pronounced when comparing WGS based score with 1000 Genomes-imputed array-based scores (r = 0.982, 0.975, 0.957 for European, South Asian, African respectively). Population-level analysis using WGS-based T1DGRS revealed significant ancestry-based stratification, with European ancestry individuals showing the highest scores, followed by South Asian (average 0.28 SD lower than Europeans, p < 10− 58) and African ancestry individuals (average 0.89 SD lower than Europeans, p < 10− 300). Notably, when applying the European ancestry-derived 90th centile risk threshold, only 0.71% (95% CI 0.41–1.13) of African ancestry individuals and 6.4% (95% CI 5.6–7.2) of South Asian individuals were identified as high-risk, substantially below the expected 10%. In conclusion, while WGS is viable for generating T1DGRS, with TOPMed-imputed genotypes offering a cost-effective alternative, the persistence of ancestry-based variations in T1DGRS distribution even using whole genome sequencing emphasises the need for ancestry-specific or pan-ancestry standards in clinical practice.https://doi.org/10.1038/s41598-024-82278-x
spellingShingle Ankit M. Arni
Diane P. Fraser
Seth A. Sharp
Richard A. Oram
Matthew B. Johnson
Michael N. Weedon
Kashyap A. Patel
Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
Scientific Reports
title Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
title_full Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
title_fullStr Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
title_full_unstemmed Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
title_short Type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array-based approaches
title_sort type 1 diabetes genetic risk score variation across ancestries using whole genome sequencing and array based approaches
url https://doi.org/10.1038/s41598-024-82278-x
work_keys_str_mv AT ankitmarni type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT dianepfraser type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT sethasharp type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT richardaoram type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT matthewbjohnson type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT michaelnweedon type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches
AT kashyapapatel type1diabetesgeneticriskscorevariationacrossancestriesusingwholegenomesequencingandarraybasedapproaches