In mytalbot/cms: Composite Measure Schemes for Severity Assessment in Animal Research

cms

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  fig.path = "man/figures/README-",
  out.width = "100%"
)

Composite measures schemes (cms)

The cms package uses tabular data from rat epilepsy studies and applies a composite measures scheme (via PCA) to select the most prominent features. Further, variables can be selected to perform cluster analysis on a subset in order to build a composite score. Finally, the cluster distribution is displayed for the subgroups and allows severity assessment between animal models.

Please note: the cms_analysis and cms_cluster functions are deprecated.

Click here for reading the cms Vignette.

Dependencies

The cms package has some dependencies. We advise installing/updating the following packages before using cms:

• ggplot2
• dplyr
• factoextra

Installation

You can install the development version from GitHub with:

# install.packages("devtools")
devtools::install_github("mytalbot/cms")
library(cms)

Example cms

The following example uses the (pre-cleaned) internalized epilepsy data (episet_full) set with three experimental subgroups. Further, the feature selection is repeated 100-fold. The example uses the new cms function. Please note that all variables that shall be included must be specified in the vars object.

library(knitr)
library(kableExtra)

Working example

Note: the following example shows the pooled data from the episet_full set, using the pooled subgroups. You might need to filter them, if you are interested in specific subsets.

library(cms)
# Do the cms feature analysis (with a limited set of variables)
usecase <- cms(raw        = episet_full,
               runs       = 100,
               idvariable = "animal_id",
               setsize    = 0.8,
               variables  = c("Sacc_pref", "social_interaction", 
                              "burrowing_rat", "openfield_rat"),
               maxPC      = 1:4,
               clusters   = 3, 
               showplot   = FALSE)

# This also shows the plot
usecase$p

Table of the cms feature frequency distributions

head(usecase$FRQ)

mytalbot/cms documentation built on July 3, 2022, 1:35 a.m.