README.md

In pdrhlik/phylomice: Multiple imputation with phylogenetic information.

phylomice

Extension of R mice package adding new imputation methods that incorporate phylogenetic information.

You should be familiar with the mice package to use this one.

Installation

You can install the package using devtools:

devtools::install_github("pdrhlik/phylomice")

Available methods

This is a list of methods that are available in the package. Each of these methods requires two additional arguments to the mice function: psi and psiinv.

• psi is a covariance matrix created from the phylogenetic tree,
• psiinv is the inverse of psi.

Both of these arguments need to be computed before running mice because of the computational cost.

Creating psi and psiinv

Let's assume that you have your tree loaded in a tree variable. You can then create psi and psiinv using the precomputePsi helper function. You need to have the ape package installed.

prec <- precomputePsi(tree)

str(prec)
List of 2
 $ psi   : ...
 $ psiinv: ...

phnorm

Imputes univariate continuous missing data using the generalized least square approach. It is based on the mice.impute.norm method in mice package.

library(phylomice)
imp <- mice(data, method = 'phnorm', psi = psi, psiinv = psiinv)

If you have precomputed psi and psiinv using precomputePsi, your call would look like this:

imp <- mice(data, method = 'phnorm', psi = prec$psi, psiinv = prec$psiinv)

phpmm

This method is currently under development.

Authors

• Patrik Drhlik - Initial work - pdrhlik
• Simon P. Blomberg - Theoretical background

If you are interested in improving current or creating new imputation methods that use phylogenies, don't hesitate to contribute.

License

This project is licensed under the GPL-3 License.

Acknowledgments

This project wouldn't be possible without the following parties: The University of Queensland, Australia Technical University of Liberec, Czech Republic * NESSIE Erasmus Mundus project

pdrhlik/phylomice documentation built on May 24, 2019, 11:45 p.m.