Boosting probabilistic graphical model inference by incorporating prior knowledge from multiple sources.

Paurush Praveen,Holger Fröhlich,Peter Csermely

doi:10.1371/journal.pone.0067410

Paurush Praveen, Holger Fröhlich + Show 1 more

Open Access

https://doi.org/10.1371/journal.pone.0067410

Copy DOI

Abstract

Inferring regulatory networks from experimental data via probabilistic graphical models is a popular framework to gain insights into biological systems. However, the inherent noise in experimental data coupled with a limited sample size reduces the performance of network reverse engineering. Prior knowledge from existing sources of biological information can address this low signal to noise problem by biasing the network inference towards biologically plausible network structures. Although integrating various sources of information is desirable, their heterogeneous nature makes this task challenging. We propose two computational methods to incorporate various information sources into a probabilistic consensus structure prior to be used in graphical model inference. Our first model, called Latent Factor Model (LFM), assumes a high degree of correlation among external information sources and reconstructs a hidden variable as a common source in a Bayesian manner. The second model, a Noisy-OR, picks up the strongest support for an interaction among information sources in a probabilistic fashion. Our extensive computational studies on KEGG signaling pathways as well as on gene expression data from breast cancer and yeast heat shock response reveal that both approaches can significantly enhance the reconstruction accuracy of Bayesian Networks compared to other competing methods as well as to the situation without any prior. Our framework allows for using diverse information sources, like pathway databases, GO terms and protein domain data, etc. and is flexible enough to integrate new sources, if available.

Highlights

Probabilistic graphical models, like (Dynamic) Bayesian Networks and Gaussian Graphical Models, have turned out to be useful for extracting meaningful biological insights from experimental data in life science research
The results showed that Bayesian Network reconstructions using Latent Factor Model (LFM) and Noisy-OR Model (NOM) priors were significantly closer to the established biological knowledge than without using any prior (Figure 5c)
We proposed two methods to integrate different, heterogenous sources of biological information in form of a consistent structure prior for probabilistic network inference

Summary

Introduction

Probabilistic graphical models, like (Dynamic) Bayesian Networks and Gaussian Graphical Models, have turned out to be useful for extracting meaningful biological insights from experimental data in life science research. Reverse engineering of regulatory network on the basis of such data is challenging and often fails to reach the desired level of accuracy. To deal with this problem one can either work at experimental level by increasing the sample size, which is practically difficult, or at the inference level by embedding biological background knowledge

Methods

Results

Conclusion