Comparative evaluation of reverse engineering gene regulatory networks with relevance networks , graphical gaussian models and bayesian networks

Motivation: An important problem in systems biology is the inference of biochemical pathways and regulatory networks from postgenomic data. Various reverse engineering methods have been proposed in the literature, and it is important to understand their relative merits and shortcomings. In the present paper, we compare the accuracy of reconstructing gene regulatory networks with three different modelling and inference paradigms: (1) Relevance networks (RNs): pairwise association scores independent of the remaining network; (2) graphical Gaussian models (GGMs): undirected graphical models with constraint-based inference, and (3) Bayesian networks (BNs): directed graphical models with score-based inference. The evaluation is carried out on the Raf pathway, a cellular signalling network describing the interaction of 11 phosphorylated proteins and phospholipids in human immunesystemcells.Weuseboth laboratorydata fromcytometry experiments as well as data simulated from the gold-standard network.We also compare passive observationswith active interventions. Results:OnGaussian observational data, BNs andGGMswere found tooutperformRNs.Thedifference inperformancewasnotsignificant for thenon-linearsimulateddataandthecytoflowdata, though.Also,wedid not observe a significant difference between BNs andGGMs on observational data in general. However, for interventional data, BNs outperformGGMsandRNs, especially when taking the edge directions rather than just theskeletonsof thegraphs intoaccount.This suggests that the higher computational costs of inference with BNs over GGMs and RNs are not justified when using only passive observations, but that active interventions in the form of gene knockouts and over-expressions are required to exploit the full potential of BNs. Availability: Data, software and supplementary material are available from http://www.bioss.sari.ac.uk/staff/adriano/research.html. Contact:adriano@bioss.ac.uk,dirk@bioss.ac.uk,Grzegorc@statistik.

[1]  K. Strimmer,et al.  Statistical Applications in Genetics and Molecular Biology A Shrinkage Approach to Large-Scale Covariance Matrix Estimation and Implications for Functional Genomics , 2011 .

[2]  Korbinian Strimmer,et al.  An empirical Bayes approach to inferring large-scale gene association networks , 2005, Bioinform..

[3]  Ming Zhou,et al.  Regulation of Raf-1 by direct feedback phosphorylation. , 2005, Molecular cell.

[4]  Satoru Miyano,et al.  Estimating gene regulatory networks and protein-protein interactions of Saccharomyces cerevisiae from multiple genome-wide data , 2005, ECCB/JBI.

[5]  Lorenz Wernisch,et al.  Reconstruction of gene networks using Bayesian learning and manipulation experiments , 2004, Bioinform..

[6]  Dirk Husmeier,et al.  Sensitivity and specificity of inferring genetic regulatory interactions from microarray experiments with dynamic Bayesian networks , 2003, Bioinform..

[7]  Satoru Miyano,et al.  Combining microarrays and biological knowledge for estimating gene networks via Bayesian networks , 2003, Computational Systems Bioinformatics. CSB2003. Proceedings of the 2003 IEEE Bioinformatics Conference. CSB2003.

[8]  Tom Burr,et al.  Causation, Prediction, and Search , 2003, Technometrics.

[9]  Satoru Miyano,et al.  Bayesian network and nonparametric heteroscedastic regression for nonlinear modeling of genetic network , 2003, Proceedings. IEEE Computer Society Bioinformatics Conference.

[10]  V. Anne Smith,et al.  Evaluating functional network inference using simulations of complex biological systems , 2002, ISMB.

[11]  Graham J. Wills,et al.  Introduction to graphical modelling , 1995 .

[12]  Nir Friedman,et al.  Inferring subnetworks from perturbed expression profiles , 2001, ISMB.

[13]  Nir Friedman,et al.  Being Bayesian about Network Structure , 2000, UAI.

[14]  Michal Linial,et al.  Using Bayesian networks to analyze expression data , 2000, RECOMB '00.

[15]  I S Kohane,et al.  Mutual information relevance networks: functional genomic clustering using pairwise entropy measurements. , 1999, Pacific Symposium on Biocomputing. Pacific Symposium on Biocomputing.

[16]  David Heckerman,et al.  A Tutorial on Learning with Bayesian Networks , 1999, Innovations in Bayesian Networks.

[17]  E. Davidson,et al.  Genomic cis-regulatory logic: experimental and computational analysis of a sea urchin gene. , 1998, Science.

[18]  David Maxwell Chickering,et al.  A Transformational Characterization of Equivalent Bayesian Network Structures , 1995, UAI.

[19]  J. York,et al.  Bayesian Graphical Models for Discrete Data , 1995 .

[20]  David Maxwell Chickering,et al.  Learning Bayesian Networks: The Combination of Knowledge and Statistical Data , 1994, Machine Learning.

[21]  David Heckerman,et al.  Learning Gaussian Networks , 1994, UAI.

[22]  I Pournara,et al.  Reconstructing gene networks by passive and active Bayesian learning. , 2005 .

[23]  Isaac S. Kohane,et al.  Relevance Networks: A First Step Toward Finding Genetic Regulatory Networks Within Microarray Data , 2003 .

[24]  Francis J. Doyle,et al.  Simulation Studies for the Identification of Genetic Networks from cDNA Array and Regulatory Activity Data , 2001 .

[25]  E. Davidson,et al.  Cis-regulatory logic in the endo16 gene: switching from a specification to a differentiation mode of control. , 2001, Development.

[26]  G. Briggs,et al.  A Note on the Kinetics of Enzyme Action. , 1925, The Biochemical journal.

[27]  M. Muir.  Physical Chemistry , 1888, Nature.