Neutral processes alone can produce extraordinary amounts of diversity — but so too can selection. How can these processes be teased apart?
If neutrality holds, only a small set of distributions can be created under given values of innovation and loss through drift. Montgomery Slatkin used this idea to develop a test for neutrality. The null model is neutrality; only if a distribution cannot be explained under this neutral model should selection be inferred.
Explore how this test behaves on datasets of genes, baby names and tree species. Note how diversity expectations change as rates of innovation and drift vary.
The input data are:
Haplotype diversity generated by a neutral drift simulation, with μ = 0.01 and N = 2000.
Y chromosome diversity in the village of Kateri on Timor. The parameters are the mutation rate and population size to model genetic data.
Frequencies of 100 random female baby names in the USA in 2015. The innovation and drift parameters are name creation and population size.
Species diversity in Bukit Timah, Singapore. Speciation and species invasions generate local diversity and the drift parameter is metacommunity size.