Multiple sclerosis (MS) is a neuroimmunological disorder of the CNS with a strong heritable component. The genetic architecture of MS susceptibility is well understood in populations of European ancestry. However, the extent to which this architecture explains MS susceptibility in populations of non-European ancestry remains unclear. In this Perspective article, we outline the scientific arguments for studying MS genetics in ancestrally diverse populations. We argue that this approach is likely to yield insights that could benefit individuals with MS from all ancestral groups. We explore the logistical and theoretical challenges that have held back this field to date and conclude that, despite these challenges, inclusion of participants of non-European ancestry in MS genetics studies will ultimately be of value to all patients with MS worldwide.[H1] The MHC locus The role of the MHC locus (6p21) in determining MS susceptibility is wellestablished 39,40 . However, studying the role of the MHC in populations of non-European ancestry is valuable because population-specific alleles exist and haplotype structures differ between populations (Figure 1). Such studies can reveal the role of HLA alleles that are not present in European populations and clarify independent and/or shared effects of alleles that highlight core disease pathways. Differences between populations of different ancestry could also reveal a role for specific pathogens or selection pressures.Determining the precise mechanisms by which MHC variation affects MS biology has been challenging owing to the density of genes in this region, the complex linkage disequilibrium, and the existence of long-range haplotypes. A GWAS in populations of European ancestry has identified 32 statistically independent signals within the MHC locus, including class II alleles that increase risk (HLA-DRB1*15:01, HLA-DRB1*03:01,HLA-DRB1*13:03, HLA-DRB1*08:01 and HLA-DQB1*03:02), class I alleles that are protective (HLA-A*02:01, HLA-B*44:02, HLA-B*38:01 and HLA-B*55:01), and some risk variants outside of classical HLA genes 8,41 . Gene-gene interactions also occur at this locus in populations of European ancestryseveral alleles modulate the effect of HLA-DRB1*15:01 8,41 . There is little evidence to suggest that the MHC influences MS phenotypes (for example, relapse rate, severity or relapsing-remitting versus progressive disease) besides age of onset in European populations [41][42][43] .The frequency, distribution and haplotypes of HLA alleles differ between ancestral populations (Figure 1) 44,45 . Some alleles are absent in the European population so their influence on MS risk cannot be studied in populations of European ancestry. Conversely, some MS risk alleles, such as HLA-DRB1*15:01, are rare in populations of non-European ancestry. The class II allele HLA-DRB1*04:05, which is essentially absent in populations of European ancestry, has been associated with MS in Japanese 46 , Turkish 47 , South American 48 , African American 10 , and Sicilian populations 49 . The combina...