- Population-specific and ancestry-related reference panels are important resources in bioinformatics, population genetics, statistical genetics, and genomic medicine. They consist of genetic data from individuals belonging to particular populations, geographic regions, or ancestry groups and are used as reference datasets for studying genetic diversity, estimating allele frequencies, inferring ancestry, imputing missing genotypes, reconstructing haplotypes, and interpreting genetic variants. These panels help researchers understand how genetic variation is distributed across human populations and improve the accuracy of genomic analyses by providing reference information that reflects the genetic backgrounds of the individuals being studied. They are particularly important because genetic variation is not distributed uniformly across the world. Allele frequencies, linkage disequilibrium patterns, haplotypes, and patterns of genetic relatedness can differ among populations as a result of demographic history, migration, genetic drift, natural selection, and admixture.
- A reference panel generally contains genotypes or genome sequences from a collection of individuals whose genetic data have been carefully processed and organized. The panel may include single-nucleotide polymorphisms (SNPs), insertions and deletions (indels), structural variants, phased haplotypes, and population or ancestry labels. A phased reference panel contains information about which alleles occur together on the same chromosome, making it particularly useful for genotype imputation and haplotype analysis. Reference panels may be designed for a broad range of populations or focus on specific geographic regions, ancestral backgrounds, or underrepresented communities. Some contain data from individuals with relatively homogeneous ancestry, while others include admixed populations with genetic contributions from multiple ancestral sources.
- The 1000 Genomes Project is one of the most widely used resources for population-specific genetic reference data. It characterized genetic variation in 2,504 individuals from 26 populations, providing a widely used catalogue of SNPs, indels, structural variants, and haplotypes. Its populations represented African, European, East Asian, South Asian, and admixed American groups. The project provides a foundation for studying population structure, allele frequencies, linkage disequilibrium, ancestry, and human evolutionary history. Its phased haplotypes have also been widely used as reference data for genotype imputation. Although the original Phase 3 dataset remains important, researchers should distinguish it from later high-coverage sequencing collections and other datasets maintained through the International Genome Sample Resource (IGSR). <Link url=”https://www.internationalgenome.org/” title=”International Genome Sample Resource”/> <Link url=”https://www.internationalgenome.org/data” title=”1000 Genomes Data”/>
- The Haplotype Reference Consortium (HRC) was developed to provide a large reference panel of phased human haplotypes, particularly for genotype imputation in European-ancestry datasets. It combined sequencing information from multiple cohorts to create a large collection of haplotypes and improved the imputation of genetic variants, especially those that are relatively uncommon in the populations represented. HRC has been influential in genome-wide association studies and population genetics, although its ancestry composition limits its suitability as a universal reference panel. A panel dominated by one ancestry may provide less accurate imputation for individuals from populations that are poorly represented. Researchers should also check the access conditions, supported genome build, and compatibility of a reference panel with their genotype data before using it. <Link url=”https://www.haplotype-reference-consortium.org/” title=”Haplotype Reference Consortium”/>
- The Trans-Omics for Precision Medicine (TOPMed) programme has generated extensive whole-genome sequencing data from diverse research cohorts, with a major emphasis on understanding genetic variation relevant to human health. TOPMed data have supported the development of reference panels that improve genotype imputation, including in populations that were less well represented in earlier panels. Its data are particularly valuable for studying rare and low-frequency variants, ancestry-related genetic variation, cardiovascular and pulmonary diseases, and other complex traits. TOPMed resources are also important for examining genetic variation beyond the common variants commonly captured by genotyping arrays. Access to individual-level TOPMed data is subject to programme-specific policies and controlled-access requirements. <Link url=”https://topmed.nhlbi.nih.gov/” title=”TOPMed Programme”/> <Link url=”https://bravo.sph.umich.edu/” title=”TOPMed BRAVO Variant Browser”/>
- The Human Genome Diversity Project (HGDP) is an important resource for studying global human genetic diversity. It includes samples from geographically diverse populations and has been used extensively to investigate population structure, human migration, genetic differentiation, and evolutionary history. The HGDP dataset has contributed to research on genetic variation in populations that were not adequately represented in early large-scale genomic studies. It has also been combined with other resources to examine ancestry, admixture, and patterns of genetic diversity. HGDP data should be interpreted with attention to sampling history, population labels, and the ethical context in which samples were collected.
- The Simons Genome Diversity Project (SGDP) provides high-quality genome sequences from individuals representing diverse populations around the world. It was developed to improve understanding of human population history and genetic diversity, including populations with relatively limited representation in other genomic datasets. SGDP has been used in research on ancient population structure, migration, admixture, genetic drift, and human evolutionary history. Together, HGDP and SGDP provide valuable reference information for studying global genetic variation, although neither should be treated as a complete or universally representative sample of all human populations. <Link url=”https://www.simonsfoundation.org/” title=”Simons Foundation”/>
- GenomeAsia 100K is a large-scale initiative designed to improve genomic representation across Asian populations. It has contributed whole-genome sequencing data from diverse Asian individuals and supports research into population-specific genetic variation, disease-associated variants, and ancestry. Asian populations are highly diverse, including many geographically and historically distinct groups. Consequently, broad categories such as East Asian, South Asian, or Southeast Asian may conceal substantial genetic differences. GenomeAsia-related data help researchers investigate this diversity and address gaps in genomic reference resources. Such datasets are particularly relevant to studies of rare variants, population history, pharmacogenomics, and precision medicine in Asian populations. <Link url=”https://www.genomeasia100k.org/” title=”GenomeAsia 100K”/>
- The African Genome Variation Project (AGVP) has contributed important information about genetic diversity across African populations. Africa contains exceptionally high levels of human genetic diversity and complex population histories, including ancient population structure, migration, admixture, and substantial variation among geographically distinct groups. AGVP data have supported studies of African population structure, genetic ancestry, disease-associated variation, and the design of more appropriate reference resources for African populations. African genomic representation remains uneven, and a single African reference panel cannot adequately represent the genetic diversity of the continent. Greater representation of geographically and ethnolinguistically diverse populations is therefore important for improving imputation, variant interpretation, and genomic medicine.
- The PAGE Study, or Population Architecture using Genomics and Epidemiology, was developed to investigate genetic architecture across diverse populations, particularly groups historically underrepresented in genomic research. It has supported the study of genetic associations, allele-frequency differences, ancestry, and disease-related variation in multiple population groups. Other research initiatives, including the Population Architecture using Genomics and Epidemiology programme, the Consortium on Asthma among African-ancestry Populations in the Americas (CAAPA), and related multi-ancestry studies, have contributed valuable data for understanding genetic variation across populations with African, Latino, Hispanic, Indigenous American, and other ancestry backgrounds. These resources are important because ancestry-related differences in allele frequencies and linkage disequilibrium can influence the transferability of genetic findings between populations. <Link url=”https://www.pagestudy.org/” title=”PAGE Study”/>
- The UK Biobank is a large population-based biomedical resource containing genetic and health-related data from hundreds of thousands of participants in the United Kingdom. It has been extensively used for genome-wide association studies, polygenic risk score development, population genetics, and genotype imputation. Its genetic data include genotyping array information and large-scale sequencing datasets. Although UK Biobank is valuable for many analyses, its participant composition does not represent every population equally. Researchers should consider its ancestry distribution and sampling characteristics when using it as a reference or comparing its results with other populations. <Link url=”https://www.ukbiobank.ac.uk/” title=”UK Biobank”/>
- The All of Us Research Program in the United States aims to build a large and diverse biomedical research resource. It includes genomic information alongside health, demographic, and lifestyle data, with an emphasis on increasing representation of communities historically underrepresented in biomedical research. Its genomic datasets can support studies of genetic diversity, ancestry, rare variants, disease associations, and precision medicine. All of Us is especially important for investigating the performance of genomic tools and risk prediction methods across diverse populations. Access to individual-level information is governed by programme-specific data access and privacy requirements. <Link url=”https://www.researchallofus.org/” title=”All of Us Research Program”/>
- Population-specific reference resources are also important for Latin American and admixed populations. Many individuals in Latin America have varying proportions of Indigenous American, European, and African ancestry, reflecting complex histories of migration, colonization, population admixture, and demographic change. The proportions and genomic patterns of these ancestry components differ considerably among countries, regions, and communities. Resources such as the Mexican Biobank and multi-ancestry studies of Latin American populations provide useful data for investigating this variation. The Mexico City Prospective Study and other regional genomic initiatives also contribute to research on population history and health, although their primary purposes and genomic coverage may differ. These resources can improve the interpretation of genetic variants and polygenic risk scores in populations that are poorly represented by European-dominated reference datasets.
- Population-specific reference panels are particularly important for genotype imputation. Genotyping arrays measure only a subset of genetic variants in an individual’s genome. Imputation uses statistical information from a reference panel of phased haplotypes to estimate the genotypes of variants that were not directly measured. The accuracy of imputation depends on factors such as reference panel size, variant frequency, linkage disequilibrium, haplotype similarity, sequencing quality, and the genetic ancestry of the study population. A reference panel containing individuals with genetic backgrounds similar to those in the study may improve imputation accuracy because it is more likely to contain relevant haplotypes and linkage disequilibrium patterns. However, ancestry labels are only approximate indicators of genetic similarity, and the best panel should be selected based on genomic compatibility rather than broad geographic labels alone. Common imputation tools include Minimac4, IMPUTE5, Beagle, and Eagle, with the appropriate tool depending on the analysis and available reference data.
- Haplotype phasing is another important application of population-specific reference panels. Phasing determines which alleles occur together on the same chromosome in diploid individuals. For example, an individual who is heterozygous at two nearby SNPs may have the alleles arranged in one of two possible haplotype configurations. Reference panels provide information about haplotypes observed in other individuals and can help infer the most likely arrangement. Accurate phasing is useful for imputation, association analysis, compound heterozygosity studies, rare-variant interpretation, and the investigation of genetic inheritance. Phasing can be affected by population structure, admixture, recombination, and the availability of closely related reference haplotypes.
- Ancestry inference is another major application of these resources. Genetic ancestry methods compare an individual’s genetic data with reference populations to estimate similarities and identify possible ancestral contributions. These methods may use principal component analysis, model-based clustering, allele-frequency comparison, haplotype-based approaches, or local ancestry inference. Principal component analysis can identify major axes of genetic variation, while tools such as ADMIXTURE estimate ancestry components under specified statistical models. Local ancestry methods examine chromosomal segments to infer their likely ancestral origins, which is particularly useful in admixed populations. Results depend strongly on the reference populations included, their sample sizes, the selected markers, and the assumptions of the statistical method. An ancestry estimate should therefore be interpreted as a model-based description of genetic similarity rather than a definitive statement about an individual’s identity, culture, nationality, or ethnicity.
- Population-specific reference panels are also essential for estimating allele frequencies. The frequency of a genetic variant can vary considerably among populations. A variant that is rare in one population may be relatively common in another because of demographic history, founder effects, genetic drift, migration, or selection. Population frequency information is important when assessing whether a variant may be relevant to a disease or trait. For example, a rare disease-associated variant should be evaluated against appropriate ancestry-related frequency data because an apparently rare variant in a poorly represented population may be more common in that population than available databases suggest. Resources such as gnomAD provide population-level allele-frequency information, but gnomAD is primarily a variant-frequency and constraint resource rather than a general-purpose phased imputation panel. Researchers should distinguish allele-frequency databases from reference panels specifically designed for phasing and imputation. <Link url=”https://gnomad.broadinstitute.org/” title=”gnomAD”/>
- The accuracy of genome-wide association studies and polygenic risk scores can also depend on the availability of ancestry-related reference data. Linkage disequilibrium patterns differ among populations, meaning that a genetic marker associated with a trait in one population may not have the same relationship with the causal variant in another. Consequently, association signals and polygenic risk scores developed mainly in one ancestry group may perform less accurately in other populations. Multi-ancestry reference panels and studies can improve the identification of shared and population-specific genetic effects, improve fine-mapping, and support more equitable genomic applications. Nevertheless, genetic ancestry is only one contributor to differences in disease risk, and environmental, social, behavioural, and healthcare-related factors must also be considered.
- Reference panels are also used in rare-variant analysis and the interpretation of clinical genetic findings. Population-specific frequency data help researchers determine whether a variant is sufficiently rare to be consistent with a suspected Mendelian disorder. However, rarity alone does not establish pathogenicity. Variant interpretation also requires evidence from functional studies, segregation analysis, phenotype compatibility, gene–disease validity, computational predictions, and established clinical classification frameworks. Resources such as gnomAD, 1000 Genomes, TOPMed, and other population datasets can provide complementary frequency information, but differences in sequencing coverage, quality control, ancestry composition, and variant-calling methods must be considered.
- The development of high-quality reference panels requires careful sampling, sequencing, genotyping, quality control, and population metadata management. Sample selection should aim to represent the genetic diversity relevant to the intended application. Genotyping or sequencing data must be checked for missingness, contamination, sample swaps, sequencing errors, and relatedness where appropriate. Variant calls should be normalized and consistently mapped to a documented reference genome assembly. For imputation panels, accurate haplotype phasing and reliable variant representation are particularly important. Population labels, sampling locations, and other metadata should be documented carefully while respecting participant consent and privacy requirements.
- An important challenge is the historical underrepresentation of many populations in genomic research. Large reference panels have often contained substantial numbers of individuals with European ancestry, while numerous African, Asian, Indigenous, Pacific Islander, and admixed populations have been less extensively represented. This imbalance can reduce imputation accuracy, limit the discovery of population-specific variants, and contribute to unequal performance of genomic prediction and clinical interpretation tools. Increasing the representation of diverse populations is therefore essential for improving the scientific quality and fairness of genomic research. However, simply increasing sample size is not sufficient; studies should also consider the diversity within broad ancestry categories and avoid treating geographically defined groups as genetically uniform populations.
- Ethical considerations are particularly important when developing and using population-specific reference panels. Genetic information can reveal sensitive information about individuals, families, and communities. Researchers must consider informed consent, privacy, data security, community engagement, appropriate governance, and restrictions on secondary data use. Indigenous and historically marginalized communities may have specific expectations regarding the ownership, interpretation, and sharing of genetic data. The CARE Principles for Indigenous Data Governance—Collective Benefit, Authority to Control, Responsibility, and Ethics—provide an important framework for responsible data practices alongside the FAIR Principles of Findability, Accessibility, Interoperability, and Reusability. Open data access should not override consent conditions or community governance requirements.
- Researchers should select reference panels according to the objectives of their analyses rather than assuming that the largest available panel is always the most appropriate. Important considerations include the ancestry composition of the study cohort, the number of individuals and haplotypes represented, the availability of rare variants, genome assembly compatibility, sequencing or genotyping quality, variant coverage, phasing quality, and access conditions. When working with admixed populations, combining suitable reference resources may be necessary, provided that differences in variant representation and data processing are addressed. Researchers should also document the panel version and release date because reference datasets and their variant content can change over time.
- Population-specific and ancestry-related reference panels provide an essential foundation for modern genomic research. Resources such as the 1000 Genomes Project, HRC, TOPMed, HGDP, SGDP, GenomeAsia 100K, UK Biobank, All of Us, and other regional and population-focused initiatives have improved the representation of human genetic diversity. By supporting genotype imputation, haplotype phasing, ancestry inference, allele-frequency estimation, genome-wide association studies, and variant interpretation, these panels help researchers investigate genetic variation more accurately. Continued development of diverse, well-documented, ethically governed reference resources is necessary to improve genomic discovery, precision medicine, and the equitable application of bioinformatics across populations.