Geographic patterns in human genetic diversity carry footprints of population history 1,2 and need to be understood to carry out global biomedicine 3,4 . Summarizing and visually representing these patterns of diversity has been a persistent goal for human geneticists [5][6][7][8][9] . However, most analytical methods to represent population structure [10][11][12][13][14] do not incorporate geography directly, and it must be considered post hoc alongside a visual summary. Here, we use a recently developed spatially explicit method to estimate "effective migration" surfaces to visualize how human genetic diversity is geographically structured (the EEMS method
15). The resulting surfaces are "rugged", which indicates the relationship between genetic and geographic distance is heterogenous and distorted as a rule. Most prominently, topographic and marine features regularly align with increased genetic differentiation (e.g. the Sahara Desert, Mediterranean Sea or Himalaya at large scales; the Adriatic, inter-island straits in near Oceania at smaller scales). We also see traces of historical migrations and boundaries of language families. These results provide visualizations of human genetic diversity that reveal local patterns of differentiation in detail and emphasize that while genetic similarity generally decays with geographic distance, there have regularly been factors that subtly distort the underlying relationship across space observed today. The fine-scale population structure depicted here is relevant to understanding complex processes of human population history and may provide insights for geographic patterning in rare variants and heritable disease risk.In many regions of the world, genetic diversity "mirrors" geography in the sense that genetic differentiation increases with geographic distance ("isolation by distance" [16][17][18] ); However, due to the complexities of geography and history, this relationship is not one of constant proportionality. The recently developed analysis method EEMS visualizes how the isolation-by-distance relationship varies across geographic space 15 Specifically, it uses a model based on a local "effective migration" rate. For several reasons, the effective migration rates inferred by EEMS do not directly represent levels of gene flow 15 ; however they are useful for conveying spatial population structure: high values of effective migration reflect genetic isolation accrues gradually with distance, and low values imply isolation accrues rapidly with distance. In turn, a map of inferred patterns of effective migration can provide a compact visualization of spatial genetic structure for large, complex samples. 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 . CC-BY-NC-ND 4.0 International license peer-reviewed) is the author/funder. It is made available under a The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/233486 doi: bioRxiv preprint first posted online Dec. ...