Skip to main content
NCBI home page
European Journal of Human Genetics logoLink to European Journal of Human Genetics
. 2025 Feb 24;33(12):1690–1697. doi: 10.1038/s41431-025-01799-9

Regional autozygosity association with albumin-to-creatinine ratio reveals a novel FTO region in an Indigenous Australian population

Vignesh Arunachalam 1, Kim N Tran 1, Wendy Hoy 2, Rodney A Lea 1, Shivashankar H Nagaraj 1,✉
PMCID: PMC12669243  PMID: 39994404

Abstract

The genetic distinctiveness of Indigenous Australian populations is well established, yet the Tiwi population remains underrepresented in genetic research. Due to their prolonged geographic isolation, these populations are prone to increased runs of homozygosity (ROH). We investigated the genetic diversity of the Tiwi population, isolated from mainland Australia for decades, based on ROH and their associations with clinical traits. We analyzed 455 whole genome sequences to identify population structure via PCA and performed a comparison with UK Biobank, Melanesian, and Polynesian cohorts. ROH assessment and genome-wide and regional measures of homozygosity were used to explore associations between clinical traits and autozygosity. Our analysis revealed distinct genetic characteristics of the Tiwi population that aligned closely with those of the Melanesian cohort. Tiwi individuals exhibited an increased burden of ROH, particularly in LINC0109, FMLN1, and RPL17P45 genes on chromosomes 2, 17, and 18, respectively, indicating prolonged isolation and genetic drift. A positive correlation was observed between genomic FROH and albumin-to-creatinine ratio (ACR) levels, suggesting a potential link between autozygosity and renal health markers. Furthermore, regional autozygosity association analysis revealed an association between elevated ACR and a region in FTO, implicating its role in obesity, kidney disease, and cardiovascular conditions. Importantly, we found that this association is strong under the recessive model. This research lays a robust foundation for further exploration of ROH mapping and its implications for disease susceptibility within Indigenous communities worldwide.

Subject terms: Genome-wide association studies, Chronic kidney disease

Background

The genetic history of Aboriginal Australians reveals a single founding population ~10–32 thousand years ago, suggesting a potential selection signal related to desert habitation [1]. Indigenous people, who make up 3.8% of the Australian population (n = 984,000) [2], exhibit high genetic distinctiveness and genetic variation [3–5]. Despite the global advancement in genomics, its application to the Indigenous population remains underexplored [6]. This is partly due to the lack of Indigenous-specific data in major databases, such as 1000 Genomes [7], the Genome Aggregation Database (gnomAD) [8], and the Human Genome Diversity Project (HGDP) [9]. Up to 34% of Indigenous-specific single nucleotide variants (SNVs) are missing from these datasets [10], and only 0.05% of Indigenous Australians are represented in genomic studies, highlighting a significant representation gap [11]. This underscores the need for more comprehensive Indigenous genomic databases to advance genomic medicine and ensure equitable healthcare [12].

Formerly known as the Sentinel Islands because of their strategic location 60 km north of Darwin in the Timor Sea, the Tiwi Islands are Australia’s largest islands after Tasmania, comprising Melville and Bathurst Islands, with a collective population of 2348 [13]. The Tiwi people lived in prolonged isolation from mainland Australia, preserving a distinct language and culture divergent from other Aboriginal Australian communities. Genomic studies reveal significant genetic differentiation in the Tiwi population due to their isolation from rising sea levels, characterized by unique genetic drift patterns, prolonged isolation, and minimal migration [10]. Our recent research highlighted a higher prevalence of kidney disease and greater heritability for urinary albumin-to-creatinine ratio (ACR) in the Tiwi and a unique kidney genetic profile compared to global populations [14].

We hypothesized that the Tiwi population exhibits a higher degree of autozygosity (FROH), as measured from runs of homozygosity (ROH). ROH are uninterrupted, identical DNA sequences arising when an individual inherits two identical haploid copies of a sequence from both parents. Smaller isolated populations typically have more ROH compared to admixed populations [15, 16]. Therefore, ROH is a valuable tool for exploring population structures and disease genetics, with insights into the history and genetic architecture of complex diseases [17]. ROH are associated with the risk of schizophrenia [18], Alzheimer’s disease [19], autism [20], coronary artery disease [21], and cognitive impairment [22].

Despite these reports, ROH and their association with disease traits have not been studied in the Tiwi population. However, recent studies on Tiwi genomes demonstrated over 10% extended homozygosity, exceeding that of Indigenous American populations threefold and Eurasian populations tenfold, likely attributable to increased background relatedness rather than consanguinity commonly observed in population isolates [10]. Notably, a previous Tiwi study speculated that there might be a potential link between kidney disease and shorter ROH. However, this study found no association between the longer ROH and ACR levels, which could be attributed to high genotyping error or low coverage rates [5]. Given the unique genetic makeup and historical context of the Tiwi population, investigating ROH patterns and their association with clinical traits is essential for understanding their relevance to population health and disease susceptibility. Previous studies conducted on the Tiwi population have not fully captured the breadth of ROH regions and their association with kidney traits, despite the high prevalence of chronic kidney conditions in this population [14, 23]. This study aims to investigate ROH prevalence, extent, and their relationship with kidney markers, establishing a foundation for ROH mapping to better understand disease susceptibility within the Tiwi community.

Methods

Study population and datasets

The whole genome sequencing (WGS) data for the Tiwi population were obtained from the blood samples of 497 Tiwi individuals collected between 2013 and 2014, representing approximately 40% of Tiwi adults [14]. This project conducted in consultation with Tiwi Elders and Indigenous research experts, adheres to NHMRC guidelines and core values of Spirit and Integrity, Cultural Continuity, Equity, Reciprocity, Respect, and Responsibility, with participants’ consent obtained for using blood samples to investigate chronic kidney disease (CKD) [5, 14] The associated phenotypic profiles include blood pressure, height, weight, waist circumference, glycated hemoglobin, serum and urine albumin, serum and urine creatinine, urinary ACR, and estimated Glomerular Filtration Rate (eGFR) (estimated using CKD-EPI 2021 equation [24]). The quality control (QC) steps were performed using Plink v1.9 which included missingness rate (--geno 0.02 –mind 0.1), Hardy-Weinberg equilibrium (p > 5*10−8), heterozygosity rate (±3 standard deviation), and minor allele frequency (--maf 0.05) [14]. Following QC, we utilized 455 samples and 4.9 million SNPs for further downstream analysis.

A total of 150,000 WGS and corresponding phenotype data were obtained from the UK Biobank (UKBB) database for comparative analyses [25]. The UKBB populations were analyzed separately according to their self-reported ethnicity which included African (n = 1320), British (n = 124,948), Caribbean (n = 1835), Chinese (n = 415), Indian (n = 1772), Irish (n = 3779), and Pakistani (n = 654) populations. We also obtained the genotyping array data for the Pacific Islands population including Polynesian and Melanesian (n = 165) [26], which captures common variants across 25 different populations; however, data from only Polynesian and Melanesian ethnicities were used for ROH analysis. The same QC steps used for the Tiwi population were applied to the UKBB to ensure consistency. We also downloaded ROH summary data from Joshi et al., which includes populations such as Amish, Hutterites (HUTT), African, European isolate, Norfolk Isolate, and Hispanic [22].

Population structure

To compare the population structures, we performed a principal component analysis (PCA) using autosomal SNPs that met QC criteria. Initially, we generated a PCA plot for the Tiwi and UKBB populations. We filtered and merged shared common variants (AF > 5%) to create a unified Plink file, followed by LD pruning with the Plink command –indep-pairwise 50 5 0.2. This step was repeated for the Polynesian and Melanesian populations alongside the Tiwi population. Due to the limited number of independent variants shared across all groups, we did not combine all populations in a single plot. Instead, we analyzed the Tiwi and UKBB populations separately from the isolated and the Tiwi populations. PCA was performed using Plink’s –pca function, and the resulting principal components (PCs) were visualized using the ggplot2 [27] package in Rv4.0.3.

ROH identification

ROH identification was performed on the common variants using the default parameters of the –homozyg option in PLINK v1.9, which calls ROH using a sliding window that scans along an individual’s SNP data to detect homozygous stretches [15, 22]. Uniform parameters were applied to detect ROH across numerous populations, including Tiwi, British, Caribbean, Chinese, Indian, Irish, Pakistani, Melanesian, and Polynesian. These datasets were used to compare the extent of ROH in the Tiwi population to other global and isolated populations.

The following measures of homozygosity were calculated in this study: the average number of ROH, the average total length of ROH, and the proportion of the autosomal genome covered by ROH (FROH). The average number of ROH is the total number of runs divided by the number of samples. The average total length of ROH is defined as the total ROH length divided by the number of samples. We used these two measures to compare ROH between the Tiwi and other populations. FROH is a measure of individual autozygosity, the ratio of the total length of ROH to the length of autosomal genomes (3 × 109) covered by SNPs. Additionally, we used Spearman’s correlation to identify the association between FROH and various traits, with non-parametric correlation accounting for the non-normality distribution of the traits. Multiple test correction was applied using the Benjamini-Hochberg (BH) method.

Regional ROH association

The Plink ROH results provided information about the ROH regions identified in all the individuals. For each sample, we extracted the ROH inclusion status for each SNP, assigning a value of 1, if the SNP was within the ROH region and 0, if not, representing regional autozygosity. We used a linear mixed model to explore the association between autozygosity estimates and traits, with phenotype as the outcome variable and autozygosity status as the independent variable [15]. This model included the top two genotype PCs, gender, and age as fixed covariates. Additionally, a genomic relationship matrix was included as a random effect to account for relationships among the individuals. We excluded samples lacking ROH regions and SNPs with a regional autozygosity frequency below 1%. The statistical significance threshold was determined using Bonferroni correction for the number of SNPs tested (n = 410k), resulting in a threshold of p = 1.21 × 10−8. Furthermore, we used Plink’s --clump function to identify independent signals by grouping correlated SNPs based on Linkage Disequilibrium (LD). We validated the genotype’s significant ROH variants and their association with ACR levels using additive and recessive models, adjusting for covariates including age, sex, and the top two genotype PCs, and performed using PLINK 2.0 [28].

Results

Tiwi population structure

The WGS data from the Tiwi population identified 19,339,732 SNVs, with 42.35% of these variants absent in the UKBB datasets. In the first PCA using the Tiwi and UKBB cohorts, PC1 and PC2 explained about 71.9% of the variance. The Tiwi cohort formed a distinct cluster within the UKBB dataset (Fig. 1a). Moreover, the Tiwi population exhibited closer associations with Chinese ethnicities compared with African, Caribbean, British, and Irish populations. In the second PCA comparing Tiwi and other isolated populations, including those of Polynesian and Melanesian from the Pacific Islands, PC1 and PC2 explained about 41.6% of the variance. Notably the Tiwi cohort remained distinct (Fig. 1b). As expected, the Tiwi Islanders clustered closer to the Melanesian population than the Polynesians. Our study utilizing UKBB data and isolated populations corroborates findings from previous investigations on the Tiwi cohort [3–5, 10] and highlights distinctive genetic attributes in the Tiwi population.

Fig. 1. Population Structure of the Tiwi population.

Fig. 1

Open in a new tab

a PCA plot of Tiwi and the UK Biobank populations (self-reported ethnicities), including African, Caribbean, Indian, Pakistani, British, Chinese, and Irish populations. PC1 and PC2 explained about 72% of the variance in the data, which consisted of 114,957 independent common SNPs among the different populations. b PCA plot of Tiwi and isolated populations, including Polynesian, and Melanesian populations. PC1 and PC2 explained about 42% of the variance in the data, which consisted of 1863 independent common SNPs among the populations.

ROH prevalence in the Tiwi population compared with other isolated populations

The Tiwi people demonstrated a significantly higher prevalence of longer ROH than general and isolated populations (Fig. 2). For instance, the average total number of ROH was 43.5 (SD = 19.3), covering an average genomic length of 135 Mb (SD = 71 Mb) (Fig. 2 and Supplementary Table S1). In comparison, the Amish and HUTT native populations exhibited an average total number of ROH of 22.6 (SD = 6.1) and 22.3 (SD = 6.3), covering an average genomic length of 113 Mb (SD = 45 Mb) and 118 Mb (SD = 53 Mb) respectively, which were significantly lower than the Tiwi population. In contrast, the Polynesian population exhibited more ROH compared with the Tiwi population. Additionally, the Tiwi individuals showed a significantly higher prevalence of ROH compared to other isolates and all ethnicities in the UKBB population (Fig. 2).

Fig. 2. Average number of ROH and average total length of ROH in various cohorts.

Fig. 2

Open in a new tab

Tiwi individuals had a higher number of ROH compared to other ethnicities present in the UK Biobank, HUTT, and Amish populations.

ROH distribution in the Tiwi population

ROH were found on every autosomal chromosome and extensively distributed across the Tiwi genomes. On average, each participant had 43.5 ROH segments >1.6 Mb, spanning a total of 135 Mb. The genome-wide SNP frequency in the ROH region is illustrated in Fig. 3, and Table 1 provides the most predominant ROH regions (≥20%). Notably, 4.0% (n = 18) of individuals had no ROH segments, 1.3% (n = 6) had shorter ROH regions (≤1.6 Mb), and the remaining 94.7% had longer ROH segments (>1.6 Mb).

Fig. 3. Distribution of Runs of Homozygosity (ROH) in the Tiwi Cohort.

Fig. 3

Open in a new tab

Manhattan plot showing the frequencies of SNPs present in ROH across the genome. The y-axis represents the frequency of SNPs present in the ROH region, and the x-axis represents the chromosomes. The blue dotted line represents a frequency of 0.20, whereas the red dotted line represents a frequency of 0.15.

Table 1.

Enriched ROH regions in the Tiwi individuals.

Chromosome Start End No. of SNPs N (%) Genes mapped
2 188129455 188285982 234 91 (20) LINC01090
17 45214912 45254071 39 96 (21.1) FMLN1
17 45254648 45287723 36 91 (20) SPATA32, MAP3K14
17 45299081 45856890 21 94 (20.6) CRHR1, SPPL2C
18 40142800 40431605 434 95 (20.8) ENSG0000286844 (uc288ffz)
Open in a new tab

Characteristics of the most frequent ROH regions in the Tiwi population (observed in ≥20% of individuals) are presented, including the chromosome number, location (start and end), number of SNPs involved, frequency, and corresponding genes within each region.

The most prevalent ROH region was on chromosome 17, followed by chromosomes 2 and 18. On chromosome 17, a region spanning 45214912–45254071 was found in 96 of 455 individuals; including 39 common SNPs mapped to the FMLN1 gene (Table 1 and Fig. 3). Another ROH region on chromosome 17 was detected in 20.0% of individuals, encompassing 21 common SNPs. Besides chromosome 17, notable ROH regions were identified on chromosome 18 (40142800–40431605) observed in 20.9% of individuals, containing 434 common SNPs in the region 18q12.3, mapped to ENSG00000286844 (uc288ffz), and chromosome 2 (188129455–188285982; 20.0% of individuals), which mapped to the long intergenic non-protein coding RNA 1090 (LINC01090) gene associated with post-traumatic stress disorder (PTSD). ROH frequencies ranging between 0.15 and 0.20 are given in Supplementary Table S2.

Association of genome-wide ROH with CKD-related phenotypes

We analyzed 15 phenotypes, including ACR, eGFR, serum creatinine, serum albumin, HbA1c, urine albumin, urine creatinine, basic anthropometrics, and blood pressure. The strongest correlation was with urinary ACR (r = 0.182, p.adj = 0.002), followed by serum albumin (r = −0.159, p.adj = 0.007). Interestingly, eGFR (r = −0.096, p.adj = 0.10) was not correlated with FROH. Additionally, glycemic control (HbA1c) had a weak correlation with FROH (r = 0.109, p.adj = 0.066), which did not withstand the BH correction (Supplementary Table S3).

Regional autozygosity association with ACR levels

We analyzed 15 phenotypes for association with regional autozygosity levels and identified one statistically significant association at the Bonferroni-corrected level (p = 1.21 × 10−8) (Fig. 4). We found that regions in chromosomes 16 and 19 were significantly associated with ACR. Increased ACR levels were significantly associated with ROH at a region of chromosome 16 that mapped to FTO and CRNDE. Additionally, increased ACR levels were significantly associated with regional autozygosity at a region of chromosome 19 that harbored members of the pregnancy-specific glycoprotein (PSG) family of genes (PSG1, PSG4, and PSG9), CD177, and PRG1.

Fig. 4. Manhattan plot of regional ROH associations with ACR genome-wide associations.

Fig. 4

Open in a new tab

The SNPs that passed the red dashed line (1.21 × 10−8) demonstrated genome-wide significance, whereas SNPs that passed the blue dashed line (1 × 10−5) demonstrated nominal significance.

In our validation using traditional GWAS, we could not replicate the associations identified through regional autozygosity analysis (Supplementary Table S5). However, many significant variants in these regions are notably more prevalent in the Tiwi population compared to the UKBB population. For instance, the SNPs in the gene CRNDE (chr16:54929371:G:T) and PSG1 (rs375340094) are either rare or absent in the UKKB population (Supplementary Table S4). The allele frequencies of these variants in the Tiwi population ranged from 9.9% to 68.2%, in contrast to 2.9% to 43.3% in the UKBB population, highlighting significant differences. Additionally, 1476 loci associated with ACR achieved genome-wide significance using the ROH approach, while 11 loci were detected using the traditional GWAS approach [14]. Interestingly, none of the genome-wide loci identified in regional autozygosity association reached either genome-wide or nominal significance using the traditional GWAS approach. These findings suggest that a recessive model may better explain ACR levels rather than the conventional additive approach.

Recessive-based SNP analysis

We applied PLINK2.0 --glm recessive function to model ACR levels in the ROH-associated region, accounting for covariates such as age, sex, and genetic PCs. We found a strong association (lowest p-value = 3.19 × 10−68) between ACR levels and SNPs in the chromosome 16 region (Supplementary Table S6), consistent with the ROH results. However, SNPs in the chromosome 19 region (lowest p-value = 8.18 × 10−7), previously identified in the ROH analysis, showed nominal association under the recessive model. Supplementary Figure S1 illustrates the effect sizes, 95% confidence intervals, and significance levels for the top independent regional autozygosity SNPs across the additive, recessive, and regional autozygosity models. These results confirm our speculation that ACR could be associated with identified regions in the recessive model.

Discussion

In this study, we examined the genetic distinctness of the Tiwi individuals through population structure and ROH analysis, and investigated the association between regional autozygosity and kidney markers. Our research significantly advances the understanding of the Tiwi population’s distinctive features, particularly concerning ROH and their association with kidney function, and represents the largest genetic study on the association between ROH and clinical phenotypes in this population. As a result, population structure analysis using PCA revealed distinct clustering patterns specific to the Tiwi population compared to the UKBB population. The PCA plot indicated that the Tiwi cohort clustered closely with the Chinese population, followed by Indian and Pakistani populations, and was distant from Caucasian and African populations (Fig. 1). These findings are consistent with those of previous studies on the Tiwi population [3–5]. In contrast, PCA between the Tiwi population and isolated populations including Polynesian and Melanesian cohorts revealed that the Tiwi population formed a distinct cluster; however, it was closest to the Melanesian population. This suggests the possibility of recent East Asian admixture in some Tiwi individuals, as indicated by the gradient shift toward the Chinese, Indian, and British populations in Fig. 1a, and the Melanesians in Fig. 2a. This likely reflects more recent admixture in the Tiwi population rather than deep shared ancestry. Notably, nearly half of the variants present in the Tiwi population were absent in the UKBB populations. These findings underscore the unique genetic profile of the Tiwi individuals and are consistent with recent research demonstrating the uniqueness of the Tiwi population within the broader spectrum of Australian Indigenous populations. Silcocks et al. highlighted that 26% of the variants observed in Australian Indigenous cohorts were absent in individuals from Papua New Guinea or in the gnomAD dataset [10]. However, further admixture analyses are needed to confirm these findings. Constructing a cohort using a similar sequencing platform would enable more overlapping SNPs, accurate comparisons, and robust admixture estimates.

The ROH analysis in the Tiwi population revealed that this population has an elevated burden of ROH. Comparative analysis revealed that Tiwi individuals exhibited a higher frequency of ROH compared to several other populations, though their frequency remained lower than that of the Polynesian cohort. In contrast, the ROH frequency in the Melanesian population was lower than that observed in the Tiwi and Polynesian populations. The Pacific Islanders, including Polynesians, Melanesians, and Micronesians, underwent prolonged periods of isolation, which fostered genetic diversity between the islands [29]. This isolation remained largely uninterrupted until the arrival of the Western population during the 16th and 19th centuries [30], potentially contributing to the elevated burden of ROH observed in the Polynesian cohort compared to the Tiwi cohort. Furthermore, the Tiwi community exhibited a greater number of ROH compared to the American native populations, including the Amish and HUTT populations. These findings corroborate a recent study on Australian Aboriginal populations, which revealed that Tiwi genomes possess more than 10% extended homozygosity compared to other Indigenous communities. Moreover, functional variations are more likely to manifest in a homozygous state in Oceania compared to other regions [10].

The most prevalent ROH region was in chromosome 17 and corresponds to FMNL1, which encodes a formin-related protein involved in morphogenesis, cytokinesis, and cell polarity [31]. According to the Human Genetic Evidence (HuGE) [32] scores from the T2D database, FMLN1 displays a robust association with weight (HuGE = 45.00), followed by a moderate association with diastolic blood pressure (9.03), and a moderate association with type 2 diabetes (7.75) [33]. Furthermore, the Kidney Tissue Atlas developed by the Kidney Precision Medicine Project (KPMP) revealed that FMNL1 exhibits elevated RNA expression in acute kidney failure and CKD patients compared to healthy individuals [34]. These findings suggest that this genomic region may play a role in the pathogenesis of various diseases, especially kidney disease and type 2 diabetes. The second most prevalent ROH region was observed in chromosome 18 and includes part of the region covered by ENSG00000286844 (uc288ffz). To the best of our knowledge, this region of the chromosome has not yet been completely characterized. Furthermore, we observed one of the longest ROH regions in chromosome 2 (2q32.3), which corresponds to LINC01090. The variants in this gene are associated with an increased risk of developing PTSD [35], and an increased risk of end-stage renal disease in type 2 diabetes [36]. These genomic regions may harbor functional variants that contribute to disorders such as PTSD and renal disease. However, most studies have focused on the European population, which highlights the need for further genetic research in Indigenous communities to confirm and validate these associations.

Here, we employed ROH mapping to identify genomic regions containing specific recessive genes and their genome-wide association with complex traits. We observed an association between elevated regional autozygosity and increased levels of urinary ACR at a specific region on chromosome 2 encompassing FTO and CRNDE. FTO is a protein-coding gene, and the variant within this gene is strongly associated with the pathogenesis of various diseases and disorders, including obesity, hypertension, and cardiovascular disorders such as heart failure, acute coronary syndrome, and aortic valve stenosis. Notably, being overweight represents a significant risk factor for CKD and is also associated with increased susceptibility to nephrolithiasis and several malignancies, including kidney cancer [37]. Another investigation revealed that FTO can modulate the m6A demethylation pathway to target mRNAs, influencing the onset and progression of cardiovascular diseases across different ethnic groups [38]. Our analysis of actual genotypes of variants in FTO based on their association with ACR levels revealed that five known independent variants exhibited a nominal association with ACR. Most importantly, the variant rs60286074 was associated with ACR (b = 0.0588, p = 0.00125) [39] in the Caucasian population. The independent variant rs55750853 was identified as a protective variant against kidney stone disease (Odds Ratio (OR) = 0.9142, p = 0.0025) [40]; however, in the Tiwi population, it was associated with increased ACR, reflecting the unique kidney profile of this population. Similarly, the variant rs118048822 was associated with an increased risk of obesity (OR = 1.25, p = 0.0018) in the Finnish (FinnGen) population [41]. These findings collectively suggest a significant association of FTO with kidney disease, hypertension, and cardiovascular disease. CRNDE is a long intergenic non-protein coding RNA gene that is transcribed into multiple transcript variants, some of which may function as non-coding RNAs [42]. The transcription of this gene is negatively regulated by insulin and insulin-like growth factors, and the gene product may regulate the expression of genes involved in metabolism [43]. The upstream variant found in this gene had a nominal association with increased ACR in the Tiwi population. Furthermore, CRNDE is also involved in biological pathways such as cell proliferation, differentiation, migration, and apoptosis, and it was found to be over-expressed in eight different types of cancers, including papillary renal cell carcinoma [44]; this led to speculation that it might be associated with kidney disease pathogenicity. In our validation using the recessive model, we identified a strong association between ACR levels and SNPs in the FTO and CRNDE gene regions. This finding reinforces our hypothesis that these regions are associated with kidney function. However, to our knowledge, we did not find any literature indicating that kidney function, particularly ACR, is predominantly recessive other than autosomal recessive polycystic kidney disease (ARPKD). This further highlights the unique genetic architecture of kidney function in the Tiwi Indigenous population.

Regional autozygosity association with ACR levels was observed in chromosome 19, encompassing members of the PSG family of genes, CD177, and PRG1. PSGs constitute a group of proteins important for pregnancy and fetal development. They are instrumental in modulating the maternal immune system’s reaction to the fetus, and are linked to immunomodulation and growth factor stimulation [45]. We also identified a nominal association between the SNPs in the PSG1 gene and ACR levels under the recessive model. CD177 is an important marker for neutrophil activation, demonstrating a strong correlation with serum levels, and it is further implicated in the severity and mortality of coronavirus disease. The evaluation of CD177 as a dependable prognostic marker holds promise for routine clinical care for coronavirus patients [46]. The p53-responsive gene PRG1 is an RNA gene predominantly associated with long non-coding RNA genes and may be involved in cellular processes regulated by p53 [31]. To the best of our knowledge, no prior studies have reported an association of this region with kidney disease/function.

Our study reveals the distinctive genetic profile of the Tiwi population, marked by a high prevalence of ROH and its association with clinical traits such as ACR. Significant links between genome-wide FROH, regional autozygosity, and elevated ACR underscore the utility of these approaches in uncovering genetic factors influencing kidney function. Regional autozygosity mapping identified key associations on chromosome 2, including FTO and CRNDE, with variants in FTO associated with increased ACR levels, especially under the recessive model. These findings suggest potential screening tools for identifying individuals at risk of kidney disease and highlight the importance of further research into ROH and its role in disease pathogenesis. However, this study has certain limitations. The data for this study were generated using different platforms, including whole-genome sequencing and SNP genotyping arrays. To ensure comparability, we applied rigorous quality control and included only common SNPs identified across populations, minimizing potential platform-specific biases, and enhancing the consistency of the results. We identified only one trait associated with FROH and regional autozygosity, likely due to limited statistical power from the sample size and the low frequency of SNPs in ROH regions, which may have constrained the robustness of our estimates. Despite these limitations, the study provides a foundational framework for future research to explore the biological significance of ROH regions, the recessive nature of kidney function, and their impact on health within Indigenous populations.

Supplementary information

Figure S1 (83.8KB, docx)
Table S1 - S6 (101.9KB, xlsx)

Acknowledgements

We thank Barry Ullungurra for his help as the key contact person with the Tiwi Islanders, Bev Mcleod and Ceri Flowers for their project management and sample and data collection, Maria Scarlett for her considerable advice and guidance on the ethics of this project, Beverley Hayhurst for the original sample collection, and, most notably, the study participants and the Tiwi Land Council for their time and ongoing support for this project. We also thank our Indigenous Advisory Committee, including Carl Nixon, Pirrawayingi Puruntatameri, and Therese Bourke, for their guidance and advice on cultural practices.

Author contributions

SN led the project. SN and RL oversaw the analysis. VA and TK designed the research. VA performed the analysis. SN and VA prepared the first draft of the manuscript. RL and WH refined the manuscript by providing critical comments. All authors read and approved the final manuscript.

Funding

Funding for this research was supported by grants from the National Health and Medical Research Council (GNT1024207) and MRFF Genomics Health Futures Mission (76757). The Centre for Chronic Disease, The University of Queensland, is supported in part by the NHMRC, Chronic Kidney Disease Centre of Research Excellence, 2016–2020 (APP1079502). The National Centre for Indigenous Genomics’ genome sequencing program is supported by grants from the Australian Genomics Health Alliance, the Australian Research Data Commons (ARDC), the Canberra Medical Society, the National Computational Infrastructure (NCI) through their ANU and National Merit Allocation Schemes, and the NHMRC (GNT1143734). Open Access funding enabled and organized by CAUL and its Member Institutions.

Data availability

The Tiwi data are stored in the QUT HPC Infrastructure in Brisbane. We have established a Data Access Advisory Committee (DAC), comprising Elders, Tiwi community members, and study investigators. Requests for external research access to Tiwi data will be evaluated by the DAC on a case-by-case basis, and access will be granted accordingly. This stringent review process is aligned with the strong recommendation from the Tiwi community and the Ethics Committee, which approved this research, to control and restrict access to Tiwi data. Requests to access the datasets should be directed to A/Prof Shiv Nagaraj. (shiv.nagaraj@qut.edu.au).

Code availability

All original code has been deposited at the GitHub repository and is publicly available from https://github.com/va1208/Tiwi_ROH_analysis.

Competing interests

The authors declare no competing interests.

Ethics approval and consent to participate

Ethical approval was formally obtained through direct consultations over 30 years with the Tiwi Land Council, Tiwi Elders, and traditional owners of the Tiwi Islands. The Tiwi Islands are privately owned and follow a permission system that recognizes the importance and value of Aboriginal responsibility toward the country and is consistent with land titles held under Australian Law. This research study followed the principles outlined in the Helsinki Declaration, and all participants provided written informed consent to participate in the study. In addition, the Northern Territory Department of Health (2012–1767), QUT HREC (No. 2022-6199-10439-HE26), and ANU HREC (No. 2014-663) provided ethical approval for this study. Ethics for human participants were reviewed by the human research ethics committees of the Northern Territory Department of Health (2012–1767), the Australian National University (2014-663), the University of Queensland (2012001146), and the University of Tasmania (H0012832).

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

The online version contains supplementary material available at 10.1038/s41431-025-01799-9.

References

  • 1.Malaspinas AS, Westaway MC, Muller C, Sousa VC, Lao O, Alves I, et al. A genomic history of Aboriginal Australia. Nature. 2016;538:207–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Australian Institute of Health and Welfare. Profile of First Nations people. 2023. Available from: https://www.aihw.gov.au/reports/australias-welfare/profile-of-indigenous-australians.
  • 3.Samarasinghe SR, Hoy W, Jadhao S, McMorran BJ, Guchelaar HJ, Nagaraj SH. The pharmacogenomic landscape of an Indigenous Australian population. Front Pharmacol. 2023;14:1180640. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Jaya Shankar A, Jadhao S, Hoy W, Foote SJ, Patel HR, Scaria V, et al. Pharmacogenomic analysis of a genetically distinct Indigenous population. Pharmacogenomics J. 2022;22:100–8. [DOI] [PubMed] [Google Scholar]
  • 5.Thomson RJ, McMorran B, Hoy W, Jose M, Whittock L, Thornton T, et al. New Genetic Loci Associated With Chronic Kidney Disease in an Indigenous Australian Population. Front Genet. 2019;10:330. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Gwynne K, Jiang S, Venema R, Christie V, Boughtwood T, Ritha M, et al. Genomics and inclusion of Indigenous peoples in high income countries. Hum Genet. 2023;142:1407–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Genomes Project C, Auton A, Brooks LD, Durbin RM, Garrison EP, Kang HM, et al. A global reference for human genetic variation. Nature. 2015;526:68–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Chen S, Francioli LC, Goodrich JK, Collins RL, Kanai M, Wang Q, et al. A genomic mutational constraint map using variation in 76,156 human genomes. Nature. 2024;625:92–100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Bergstrom A, McCarthy SA, Hui R, Almarri MA, Ayub Q, Danecek P, et al. Insights into human genetic variation and population history from 929 diverse genomes. Science. 2020;367:eaay5012. [DOI] [PMC free article] [PubMed]
  • 10.Silcocks M, Farlow A, Hermes A, Tsambos G, Patel HR, Huebner S, et al. Indigenous Australian genomes show deep structure and rich novel variation. Nature. 2023;624:593–601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Popejoy AB, Fullerton SM. Genomics is falling on diversity. Nature. 2016;538:161–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Nagaraj SH, Toombs M. The Gene-Drug Duality: Exploring the Pharmacogenomics of Indigenous Populations. Front Genet. 2021;12:687116. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Tiwi Islands 2021 Census All persons QuickStats: Australian Bureau of Statistics; 2021. [Available from: https://www.abs.gov.au/census/find-census-data/quickstats/2021/702031060.
  • 14.Arunachalam V, Lea R, Hoy W, Lee S, Mott S, Savige J, et al. Novel genetic markers for chronic kidney disease in a geographically isolated population of Indigenous Australians: Individual and multiple phenotype genome-wide association study. Genome Med. 2024;16:29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lynch MT, Maloney KA, Xu H, Perry JA, Center RG, Shuldiner AR, et al. Associations of genome-wide and regional autozygosity with 96 complex traits in old order Amish. BMC Genomics. 2023;24:134. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Fieder M, Mitchell BL, Gordon S, Huber S, Martin NG. Ethnic Identity and Genome Wide Runs of Homozygosity. Behav Genet. 2021;51:405–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ceballos FC, Hazelhurst S, Ramsay M. Assessing runs of Homozygosity: a comparison of SNP Array and whole genome sequence low coverage data. BMC Genomics. 2018;19:106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Keller MC, Simonson MA, Ripke S, Neale BM, Gejman PV, Howrigan DP, et al. Runs of homozygosity implicate autozygosity as a schizophrenia risk factor. PLoS Genet. 2012;8:e1002656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ghani M, Reitz C, Cheng R, Vardarajan BN, Jun G, Sato C, et al. Association of Long Runs of Homozygosity With Alzheimer Disease Among African American Individuals. JAMA Neurol. 2015;72:1313–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Lin PI, Kuo PH, Chen CH, Wu JY, Gau SS, Wu YY, et al. Runs of homozygosity associated with speech delay in autism in a taiwanese han population: evidence for the recessive model. PLoS One. 2013;8:e72056. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Christofidou P, Nelson CP, Nikpay M, Qu L, Li M, Loley C, et al. Runs of Homozygosity: Association with Coronary Artery Disease and Gene Expression in Monocytes and Macrophages. Am J Hum Genet. 2015;97:228–37. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Joshi PK, Esko T, Mattsson H, Eklund N, Gandin I, Nutile T, et al. Directional dominance on stature and cognition in diverse human populations. Nature. 2015;523:459–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Hoy WE, Mott SA, McDonald SP. An update on chronic kidney disease in Aboriginal Australians. Clin Nephrol. 2020;93:124–8. [DOI] [PubMed] [Google Scholar]
  • 24.National Kidney Foundation: CKD-EPI creatinine Equation. 2021. [Available from: https://www.kidney.org/professionals/kdoqi/gfr_calculator/formula.
  • 25.Halldorsson BV, Eggertsson HP, Moore KHS, Hauswedell H, Eiriksson O, Ulfarsson MO, et al. The sequences of 150,119 genomes in the UK Biobank. Nature. 2022;607:732–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Ioannidis AG, Blanco-Portillo J, Sandoval K, Hagelberg E, Miquel-Poblete JF, Moreno-Mayar JV, et al. Native American gene flow into Polynesia predating Easter Island settlement. Nature. 2020;583:572–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Wickham H. ggplot2: Elegant Graphics for Data Analysis. New York: Spring-Verlag; 2016.
  • 28.Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience. 2015;4:7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Friedlaender JS, Friedlaender FR, Reed FA, Kidd KK, Kidd JR, Chambers GK, et al. The genetic structure of Pacific Islanders. PLoS Genet. 2008;4:e19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Penman BS, Gupta S, Shanks GD. Rapid mortality transition of Pacific Islands in the 19th century. Epidemiol Infect. 2017;145:1–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Stelzer G, Rosen N, Plaschkes I, Zimmerman S, Twik M, Fishilevich S, et al. The GeneCards Suite: From Gene Data Mining to Disease Genome Sequence Analyses. Curr Protoc Bioinformatics. 2016;54:30 1–1 3. [DOI] [PubMed] [Google Scholar]
  • 32.Dornbos P, Singh P, Jang DK, Mahajan A, Biddinger SB, Rotter JI, et al. Evaluating human genetic support for hypothesized metabolic disease genes. Cell Metab. 2022;34:661–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Costanzo MC, von Grotthuss M, Massung J, Jang D, Caulkins L, Koesterer R, et al. The Type 2 Diabetes Knowledge Portal: An open access genetic resource dedicated to type 2 diabetes and related traits. Cell Metab. 2023;35:695–710 e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Lake BB, Menon R, Winfree S, Hu Q, Melo Ferreira R, Kalhor K, et al. An atlas of healthy and injured cell states and niches in the human kidney. Nature. 2023;619:585–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Guffanti G, Galea S, Yan L, Roberts AL, Solovieff N, Aiello AE, et al. Genome-wide association study implicates a novel RNA gene, the lincRNA AC068718.1, as a risk factor for post-traumatic stress disorder in women. Psychoneuroendocrinology. 2013;38:3029–38. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.van Zuydam NR, Ahlqvist E, Sandholm N, Deshmukh H, Rayner NW, Abdalla M, et al. A Genome-Wide Association Study of Diabetic Kidney Disease in Subjects With Type 2 Diabetes. Diabetes. 2018;67:1414–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Kovesdy CP, Furth SL, Zoccali C. World Kidney Day Steering C. Obesity and Kidney Disease: Hidden Consequences of the Epidemic. Can J Kidney Health Dis. 2017;4:2054358117698669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Xu ZY, Jing X, Xiong XD. Emerging Role and Mechanism of the FTO Gene in Cardiovascular Diseases. Biomolecules. 2023;13:850. [DOI] [PMC free article] [PubMed]
  • 39.Teumer A, Li Y, Ghasemi S, Prins BP, Wuttke M, Hermle T, et al. Genome-wide association meta-analyses and fine-mapping elucidate pathways influencing albuminuria. Nat Commun. 2019;10:4130. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Hao X, Shao Z, Zhang N, Jiang M, Cao X, Li S, et al. Integrative genome-wide analyses identify novel loci associated with kidney stones and provide insights into its genetic architecture. Nat Commun. 2023;14:7498. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Kurki MI, Karjalainen J, Palta P, Sipilä TP, Kristiansson K, Donner KM, et al. FinnGen provides genetic insights from a well-phenotyped isolated population. Nature. 2023;613:508–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Maglott D, Ostell J, Pruitt KD, Tatusova T. Entrez Gene: gene-centered information at NCBI. Nucleic Acids Res. 2005;33:D54–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.O’Leary NA, Wright MW, Brister JR, Ciufo S, Haddad D, McVeigh R, et al. Reference sequence (RefSeq) database at NCBI: current status, taxonomic expansion, and functional annotation. Nucleic Acids Res. 2016;44:D733–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Ma X, Jin W, Zhao C, Wang X, Wang K. CRNDE: A valuable long noncoding RNA for diagnosis and therapy of solid and hematological malignancies. Mol Ther Nucleic Acids. 2022;28:190–201. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Moore T, Williams JM, Becerra-Rodriguez MA, Dunne M, Kammerer R, Dveksler G. Pregnancy-specific glycoproteins: evolution, expression, functions and disease associations. Reproduction. 2022;163:R11–R23. [DOI] [PubMed] [Google Scholar]
  • 46.Levy Y, Wiedemann A, Hejblum BP, Durand M, Lefebvre C, Surenaud M, et al. CD177, a specific marker of neutrophil activation, is associated with coronavirus disease 2019 severity and death. iScience. 2021;24:102711. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1 (83.8KB, docx)
Table S1 - S6 (101.9KB, xlsx)

Data Availability Statement

The Tiwi data are stored in the QUT HPC Infrastructure in Brisbane. We have established a Data Access Advisory Committee (DAC), comprising Elders, Tiwi community members, and study investigators. Requests for external research access to Tiwi data will be evaluated by the DAC on a case-by-case basis, and access will be granted accordingly. This stringent review process is aligned with the strong recommendation from the Tiwi community and the Ethics Committee, which approved this research, to control and restrict access to Tiwi data. Requests to access the datasets should be directed to A/Prof Shiv Nagaraj. (shiv.nagaraj@qut.edu.au).

All original code has been deposited at the GitHub repository and is publicly available from https://github.com/va1208/Tiwi_ROH_analysis.

Articles from European Journal of Human Genetics are provided here courtesy of Nature Publishing Group

RESOURCES