Potential risks and solutions for sharing genome summary data from African populations

Abstract Genome data from African population can substantially assist the global effort to identify aetiological genetic variants, but open access to aggregated genomic data from these populations poses some significant risks of community- and population- level harms. A recent amendment to National Institutes of Health policy, following various engagements with predominantly North American scientists, requires that genomic summary results must be made available openly on the internet without access oversight or controls. The policy does recognise that some sensitive, identifiable population groups might be harmed by such exposure of their data, and allows for exemption in these cases. African populations have a very wide and complex genomic landscape, and because of this diversity, individual African populations may be uniquely re-identified by their genomic profiles and genome summary data. Given this identifiability, combined with additional vulnerabilities such as poor access to health care, socioeconomic challenges and the risk of ethnic discrimination, it would be prudent for the National Institutes of Health to recognise the potential of their current policy for community harms to Africans; and to exempt all African populations as sensitive or vulnerable populations with regard to the unregulated exposure of their genome summary data online. Three risk-mitigating mechanisms for sharing genome summary results from African populations to inform global genomic health research are proposed here; namely use of the Beacon Protocol developed by the Global Alliance for Genomics and Health, user access control through the planned African Genome Variation Database, and regional aggregation of population data to protect individual African populations from re-identification and associated harms