作者: JOHN H. GILLESPIE
DOI: 10.1038/266443A0
关键词:
摘要: A MAJOR effort in theoretical population genetics has been aimed at developing statistical tests to distinguish between various explanations of observed patterns enzyme variation1. Perhaps the most useful result emerge from this is Ewens' sampling theory neutral alleles2. This allows one use number and configuration alleles a sample test for departures neutrality. Watterson3 derived analogous distribution finite subject joint effects mutation symmetric heterotic selection. It differs theory, providing stronger hypothesis-testing situation than previously available. In note I present third distribution. applies an infinite which selective coefficients fluctuate random through time space. Quite unexpectedly, turns out be identical that model, indicating agreement data with does not rule possibility natural selection responsible genetic variation sample. Furthermore, properties random-selection populations equilibrium are indistinguishable neutral-allele observations made single point. calls into question claim predictions model argues against as important factor maintenance variation.