DeepAI AI Chat
Log In Sign Up

Accounting for missing actors in interaction network inference from abundance data

07/28/2020
by   Raphaëlle Momal, et al.
AgroParisTech
0

Network inference aims at unraveling the dependency structure relating jointly observed variables. Graphical models provide a general framework to distinguish between marginal and conditional dependency. Unobserved variables (missing actors) may induce apparent conditional dependencies.In the context of count data, we introduce a mixture of Poisson log-normal distributions with tree-shaped graphical models, to recover the dependency structure, including missing actors. We design a variational EM algorithm and assess its performance on synthetic data. We demonstrate the ability of our approach to recover environmental drivers on two ecological datasets. The corresponding R package is available from github.com/Rmomal/nestor.

READ FULL TEXT
07/20/2018

Learning the effect of latent variables in Gaussian Graphical models with unobserved variables

The edge structure of the graph defining an undirected graphical model d...
06/13/2012

Adaptive Inference on General Graphical Models

Many algorithms and applications involve repeatedly solving variations o...
02/27/2023

Mixtures of All Trees

Tree-shaped graphical models are widely used for their tractability. How...
01/16/2014

Nonparametric Latent Tree Graphical Models: Inference, Estimation, and Structure Learning

Tree structured graphical models are powerful at expressing long range o...
01/31/2014

Marginal and simultaneous predictive classification using stratified graphical models

An inductive probabilistic classification rule must generally obey the p...
10/02/2017

Detecting Epistatic Selection with Partially Observed Genotype Data Using Copula Graphical Models

Recombinant Inbred Lines derived from divergent parental lines can displ...
11/15/2016

Unsupervised Learning with Truncated Gaussian Graphical Models

Gaussian graphical models (GGMs) are widely used for statistical modelin...

Code Repositories

nestor

Meet nestor, an R package for the variational inference of species interaction networks from abundance data, while accounting for missing actors.


view repo