Microbiome Differential Abundance Simulation
================
[](http://bioconductor.org/checkResults/release/bioc-LATEST/microbiomeDASim/)
[](https://bioconductor.org/packages/microbiomeDASim)
[](http://bioconductor.org/packages/stats/bioc/microbiomeDASim/)
This package is developed to simulate microbiome data for longitudinal
differential abundance analyses. Microbiome data have a variety of
features that make typical simulation methods inappropriate. For an
in-depth description of the types of problems this simulation package is
designed to solve, plus basics of the functionality please refer to the
F1000 manuscript
[f1000research.20660.2](https://doi.org/10.12688/f1000research.20660.2).
## Installation
To install the `microbiomeDASim` package the latest release version is
available from Bioconductor
``` r
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("microbiomeDASim")
```
Alternatively to use the latest development version from Bioconductor
use the following commands
``` r
if (!requireNamespace("BiocManager", quietly = TRUE))
install.packages("BiocManager")
BiocManager::install("microbiomeDASim", version="devel")
```
## Examples
An interactive examples for how to simulate data from a multivariate
normal distribution and fit a trend line using
[metagenomeSeq::fitTimeSeries](https://bioconductor.org/packages/release/bioc/html/metagenomeSeq.html)
are available in the `inst/scripts` directory.
This notebook can be run interactively using Google Collab by clicking
the “Open in Colab” marker at the top of the notebook.