Skip to content
Snippets Groups Projects
README.md 17.27 KiB

PLNmodels: Poisson lognormal models

The Poisson lognormal model and variants can be used for analysis of mutivariate count data. This package implements efficient algorithms to fit such models.

Installation

PLNmodels is available on pypi. The development version is available on Gitlab.

R Package installation

Usage and main fitting functions

Description of the package

The package implements 2 differents classes that fits a Poisson-Log-Normal model.

  • The PLN class fits a PLN model with full covariance matrix.
  • The IMPS_PLN fits a PLN-PCA model using Importance sampling.

IMPS_PLN is always slower than fastPLN. fastPLNPCA is faster than fastPLN only for datasets with very large number of genes (p>5000, see here). However, fastPLNPCA is convenient since it allows to get the Principal Components (PCs) directly, in contrary to fastPLN. To get the PCs using fastPLN, you first need to fit the model and do a PCA on the matrix \Sigma found. The numerical complexity is always linear with respect to the number of samples n_samples (see here)