Wright's hierarchical F-statistics.

This Perspective article offers a meditation on FST and other quantities developed by Sewall Wright to describe population structure, defined as any departure from reproduction through random union of gametes. Concepts related to the F-statistics draw from studies of the partitioning of variation, identity coefficients, and diversity measures. Relationships between the first two approaches have recently been clarified and unified. This essay addresses the third pillar of the discussion: Nei's GST and related measures. A hierarchy of probabilities of identity-by-state (IBS) provides a description of the relationships among levels of a structured population with respect to genetic diversity. Explicit expressions for the IBS probabilities are determined for models of structured populations undergoing regular inbreeding and recurrent mutation. Levels of genetic diversity within and between subpopulations reflect mutation as well as migration. Accordingly, indices of population structure are inherently locus-specific, contrary to the intentions of Wright. Some implications of this locus-specificity are explored.