Two-epoch Demographic History Modeling with dadi

This app uses dadi to simulate two-epoch demographic histories for comparison with population genomic data.

The two-epoch model posits that the population size changed instantaneously at some time in the past.
The parameter Na is the ancestral effective population size. The parameter Nc is the contemporary effective population size. The parameter Tgen is number of generations in the past at which the size change happened.

Use the sliders to adjust the values of Na, Nc, and Tgen. Note that the slides adjust the log10 values of the parameters, because they can vary over orders of magnitude in real data. Notice how the allele frequency spectrum changes. (Parameterizations with small Nc/Na and large Tgen can take a long time to compute.)

The Na, Nc, and Tgen fields report the non-log10 values of those parameters. Also plotted is a demesdraw representation of the demographic model.

You can upload a file containing population genomic data to compare your models with. (First, click the "X" button to clear the Data file field.) If you upload a VCF file (.vcf or .vcf.gz), the app will lump all samples in that file into a single population to analyze. (The dadi VCF parser is not fast, so be patient when loading VCF data.) You can also upload a single-population dadi frequency spectrum (.fs) file. The preloaded data are from the 2024 GHIST bottleneck challenge.

You can also download a YAML file with your current parameter settings, for easy submission to the GHIST bottleneck demographic history challenges.

Questions? Contact Ryan Gutenkunst at rgutenk@arizona.edu.
To learn more about research from the Gutenkusnt group, visit our webpage.
Development of this app was supported by NIH NIGMS grant R35 GM149235 to Ryan Gutenkunst.