CMIMN: Conditional Mutual Information for constructing Microbiome Network

CMIMN is an R package designed to construct microbiome networks using conditional mutual information (CMI) to identify relationships between taxa. The package applies two levels of CMI computation, using quantile thresholds to filter edges in the resulting adjacency matrices, facilitating robust microbial interaction analysis.

Installation
Running CMIMN Package
Example

Installation

install.packages("devtools")
devtools::install_github("solislemuslab/CMIMN")

Running CMIMN Package

loading the Data

We use the American Gut data from SpiecEasi package to run CMiNet algorithm to construct consensus microbiome network. First, load CMIMN and the American Gut Project data (included with the SpiecEasi package), which is automatically loaded alongside CMIMN).

library(CMIMN)
data = amgut1.filt

Parameters

data: A numeric matrix where rows represent taxa and columns represent samples. If quantitative is TRUE, data will be log-transformed.
q1: A numeric value representing the quantile threshold for filtering edges in the order 0 adjacency matrix.
q2: A numeric value representing the quantile threshold for filtering edges in the order 1 adjacency matrix.
quantitative: A logical value indicating if the data is quantitative. If TRUE, the data is log-transformed.

Return Value

The function returns a list with the following components:

G_order0: The adjacency matrix after order 0 calculation.
G_order1: The adjacency matrix after order 1 calculation.
Gval_order0: The CMI values for each taxa pair in the order 0 adjacency matrix.
Gval_order1: The CMI values for each taxa pair in the order 1 adjacency matrix.
quantile_order0: The quantile threshold used for filtering the order 0 adjacency matrix.
quantile_order1: The quantile threshold used for filtering the order 1 adjacency matrix.
sum_order0: The sum of edges in the order 0 adjacency matrix.
sum_order1: The sum of edges in the order 1 adjacency matrix.

Example

result <- conditional_MI( data ,q1 = 0.7, q2 = 0.95,quantitative = TRUE)

Reporting Issues and Asking Questions

If you encounter a bug, experience a failed function, or have a feature request, please open an issue in the GitHub issue tracker.

License

Citation

If you use CMIMN in your work, we kindly ask that you cite the following paper:

@article{aghdam2024,
  year = {2024},
  publisher = {In process},
  author = {Rosa Aghdam, Shan Shan, Richard Lankau and Claudia Solis-Lemus},
  title = {Leveraging Machine Learning and Enhanced Network-based methods in Potato Disease Interactions}
}
@article{aghdam2024_2,
  year = {2024},
  publisher = {In process},
  author = {Rosa Aghdam and Claudia Solis-Lemus},
  title = {CMiNet: R package for learning the Consensus Microbiome Network}
}

solislemuslab/CMIMN