This repository contains all the materials that are necesary to reproduce the figures and tables of the paper, including a set of novel predictions.
There are two main sections: the simulations folder and the parameter-estimates folder. The first deals with the simulation study generating random annotated phylogenetic trees using the PANTHER database, the latter fits the pooled-data model using 138 different sets of annotations and makes the predictions, all using a combination between the GO dataset and the PANTHER phylogenetic trees.
All of the methods presented in this paper are available in the R package
aphylo. To install the aphylo
package, you should use the following command:
devtools::install_github("USCbiostats/aphylo")In general, each dataset, figure, or table has its own R script used to be generated.
Furthermore, resulting files have the same name of the R script that created it, only
changing in the extension, for example, the R script candidate_trees.r generates the file
candidate_trees.rds, which has the 138 phylogenetic trees (including annotations) used
throughtout the paper.
-
data-raw Contains raw data used in the paper. The main dataset here is the set of experimental annotations from GOA.
-
data Contains the scripts used to process the data, including, reading the panther trees, GOA annotations, and combining them into
aphylo_treeobjects. This also contains the resulting data. -
fig Most figures of the paper, including the code used to generate the two trees featured in the paper (the low and high MAE).
-
parameter-estimates All the code used to fit the pooled models, including the obtained paramter estimates.
-
proposed-annotations Code used to generate annotation proposals on genes with no experimental annotations. The folder includes the actual table with proposed annotations.
-
sifter Code used to analyze SIFTER data.
-
simulations Code to generate the large simulation study in which we analyze the properties of MCMC and MLE estimates.