An African genotype imputation service using a diverse African population reference panel

    H3Africa/H3ABioNet Imputation Service

    A genotype phasing and imputation service for inferring unobserved genotypes in a study population based on known genotypes from a reference panel. The H3Africa/H3ABioNet Imputation Service is accessible at and provided by H3ABioNet in collaboration with H3Africa and other research projects in Africa.

    For more information on the imputation service, please log a ticket with the H3ABioNet Helpdesk.

    Why do we need an African imputation service?

    An African genotype imputation service, equipped with a comprehensive African reference panel, is indispensable for accurate genetic analyses tailored to the continent's diverse genetic landscape.

    These curated reference panels are built using genomic data sourced from the H3Africa consortium and other African initiatives, following established guidelines and workflows.

    This service will also help address challenges of data privacy, requiring genetic data to remain within the continent's boundaries, safeguarding sensitive information and respecting regulatory frameworks.

    How does it work?

    1. Create an account on the imputation service at
    2. Prepare your GWAS data in VCF format.
    3. Upload your data
    4. The imputation service will run the imputation (Minimac4) and phasing (Eagle2).
    5. Download phased/imputed files in VCF format.

    Who should use this service? 

    Researchers seeking to impute extensive GWAS samples against a rich African reference panel in a standardized manner will find this service particularly beneficial. Leveraging the same software and workflow as the Michigan Imputation Service and TopMed Imputation Service, ensures familiarity and confidence in genetic imputation procedures.



    Reference Panels

    Currently, the service only offers 1000 Genomes Phase 3 and H3Africa reference panels. To get access to the H3Africa reference panel, please log a ticket with the H3ABioNet Helpdesk.


    H3AFRICA v6 (BUILD 38)


    A High-quality reference panel of 8,894 high-coverage haplotypes from 48 populations worldwide was generated, including 50% of African ancestry.

    Reference panel information
    Biallelic SNPs

    Sample size by population

    1. Algeria (28)
    2. Benin (50)
    3. Botswana (54)
    4. Burkina Faso (33)
    5. Cameroon (50)
    6. Central African Republic (26)
    7. DRC (13)
    8. Egypt (10)
    9. Gambia (115)
    10. Ghana (26)
    11. Guinea (20)
    12. Kenya (108)
    13. Mali (50)
    14. Namibia (7)
    15. Nigeria (727)
    16. Senegal (2)
    17. Sierra Leone (87)
    18. South Africa (271)
    19. Sudan (2)
    20. Tanzania (10)
    21. Western Sahara (2)
    22. Zambia (41)
    1. Finland (99)
    2. France (46)
    3. Italy (147)
    4. Russia (61)
    5. Spain (107)
    6. United Kingdom (104)
    1. Afghanistan (15)
    2. Bangladesh (86)
    3. Cambodia (8)
    4. China (439)
    5. India (205)
    6. Israel (42)
    7. Japan (131)
    8. Myanmar (5)
    9. Pakistan (251)
    10. Palestine (83)
    11. Sri Lanka (102)
    12. Vietnam (99)
    1. Barbados (96)
    2. Brazil (15)
    3. Colombia (99)
    4. Mexico (31)
    5. Puerto Rico (104)
    6. United States (224)
    7. Peru (85)
    1. Papua New Guinea (25)
    1000 Genomes Phase 3 (Version 5)



    How to Cite

    We encourage researchers who utilize the Imputation Server for their GWAS studies to cite the following resources:




    • We sincerely thank all study investigators who contributed their valuable data to the reference panel. Their generosity and willingness to share these data, now accessible through this service, have been instrumental in its development and success.
    • We recognize and appreciate the significant contributions of the H3ABioNet Refimpute project members in generating the final reference panel and associated pipelines.
    • The project is governed by the University of Cape Town Human Research Ethics Committee, Approval No: HREC R043/2018.
    • Finally, we extend our profound gratitude to the participants who generously provided biological samples.
    © 2021
    Terms - Privacy - Policy & Safety - Helpdesk
    The H3ABioNet website content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Health