Public data for use with ASMC.
Files in this repository can be used to run examples for ASMC and FastSMC, and is also used internally for regression testing.
Two sets of decoding quantities with corresponding CSFS files and intervals information.
Both sets were generated with the CEU.demo demography in 📁 /demographies, the correspondingly-named discretizations in 📁 /discretizations, and frequencies information from UKBB in 📁 /frequencies.
The 10-20-2000 uses a finer time discretization with steps of size 10.0 and 20.0 up to 2000.0, while 30-100-2000 uses a coarser time discretization with steps of size 30.0 and 100.0 up to 2000.0.
Contains the CEU demographic model used to generate the discretizations and decoding quantities, along with all other demographic models that can be used to prepare decoding quantities by using their corresponding three letter code.
These demographies are haploid.
These models were inferred using smc++ in the PyRho paper and corresponds to these population sizes, but rescaled to assume mutation rate of 1.65e-8.
Contains the discretizations used to generate the decoding quantities.
As explained above, the 10-20-2000 is a finer time discretization (159 points) than 30-100-2000 (69 points).
Both use manually-specified time values up to 2000.0, after which a number of quantiles are generated from the demography using an exponential model.
Frequencies information from UKBB.