The goal of AwesomePackage is to infer ancestry with PSD model and fit PSD model with some algorithms.
We use the classical PSD model for ancestor inference, which has been widely used, such as STRUCTURE (Pritchard et al. 2000, MCMC), FRAPPE (Tang et al. 2005, EM), ADMIXTURE (Alexander et al. 2009, SQP), fastSTRUCTURE (Raj et al. 2014, VI), TeraStructure (Gopalan et al. 2017, SVI). We illustrate the close relationship between the PSD model, the Poisson NMF model, the multinomial topic model and the LDA model, which can optimize the algorithm. We use Expectation-Maximization algorithm (EM), sequential quadratic programming algorithm (SQP), variational inference algorithm (VI) and stochastic variational inference algorithm (SVI) to fit the model, then illustrate the relationships and differences between these algorithms through simulation experiments and real data experiments.
Installation
You can install the development version of AwesomePackage from GitHub with:
# install.packages("devtools")
devtools::install_github("JONATHONCHOW/AwesomePackage")Example
You can check out theories and examples at Articles in AwesomePackage.
Quick start
You can use the following code to see if AwesomePackage has been successfully installed.
library(AwesomePackage)
hello_world()
#> [1] "Data science is fantastic!"You can refer to Reference in AwesomePackage for the use of functions, and then you can have fun with ancestry inference!