Description Usage Arguments Details Value Author(s) Examples
This command provides a way to test for differences in event count profiles among two groups of reconstructed cophylogenies.
1 | DEEPR(group_1, group_2, perm_number = 999)
|
group_1 |
A matrix of event counts for the first group of cophylogenies. Rows represent individual cophylogenies. Columns represent counts for specific event types. |
group_2 |
A matrix of event counts for the second group of cophylogenies. Rows represent individual cophylogenies. Columns represent counts for specific event types. |
perm_number |
The number of permutations to be performed. By default 999 permutations are performed. |
A Dirichlet-multinomial model is fit to each of the groups, such that the two groups can be described by their respective model parameters, and a Log-Likelihood Ratio Test (LLRT) can be performed to compare the two models. A permutation test is performed by repeatedly generating new groups by random sampling of the originals, without replacement. As with the original two groups, the LLRT statistic is calculated for each permutation. In doing so, a distribution or LLRT values is obtained and a p-value is estimated.
p_value |
The p value of the test. |
group_1_pi |
Estimated event probabilities for group 1 |
group_2_pi |
Estimated event probabilities for group 2 |
group_1_alphas |
Estimated Dirichlet parameters for group 1 |
group_2_alphas |
Estimated Dirichlet parameters for group 2 |
group_1_theta |
Estimated overdispersion parameter for group 1 |
group_1_theta |
Estimated overdispersion parameter for group 2 |
The package was written by Mark T Merilo, with contributions to the statistical methodology and biological considerations from R. Ayesha Ali, and Sarah J. Adamowicz.
1 2 3 4 5 6 7 8 9 | #Specify the cophylogeny event datasets to use
#In this example event counts are artificially generated
groupA<-simCophy(c(1,1,1,1), 20)
groupB<-simCophy(c(1,1,1,5), 20)
#Compare cophylogeny groups
#Here an extremely small permutation number is used here to shorten example run time
#A permutation number of 999 is recommended for real analysis.
DEEPR(groupA,groupB,9)
|
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.