No retraction, correction or concern found when last checked on Oct 27, 2026.

Regularized estimation of large-scale gene association networks using graphical Gaussian models

Nicole Krämer, Juliane Schäfer, Anne-Laure Boulesteix

BMC Bioinformatics · 2009

Abstract

Background. Graphical Gaussian models are popular tools for the estimation of (undirected) gene association networks from microarray data. A key issue when the number of variables greatly exceeds the number of samples is the estimation of the matrix of partial correlations. Since the (Moore-Penrose) inverse of the sample covariance matrix leads to poor estimates in this scenario, standard methods are inappropriate and adequate regularization techniques are needed. Popular approaches include biased estimates of the covariance matrix and high-dimensional regression schemes, such as the Lasso and Partial Least Squares.

Results. In this article, we investigate a general framework for combining regularized regression methods with the estimation of Graphical Gaussian models. This framework includes various existing methods as well as two new approaches based on ridge regression and adaptive lasso, respectively. These methods are extensively compared both qualitatively and quantitatively within a simulation study and through an application to six diverse real data sets. In addition, all proposed algorithms are implemented in the R package "parcor", available from the R repository CRAN.

Conclusion. In our simulation studies, the investigated non-sparse regression methods, i.e. Ridge Regression and Partial Least Squares, exhibit rather conservative behavior when combined with (local) false discovery rate multiple testing in order to decide whether or not an edge is present in the network. For networks with higher densities, the difference in performance of the methods decreases. For sparse networks, we confirm the Lasso's well known tendency towards selecting too many edges, whereas the two-stage adaptive Lasso is an interesting alternative that provides sparser solutions. In our simulations, both sparse and non-sparse methods are able to reconstruct networks with cluster structures. On six real data sets, we also clearly distinguish the results obtained using the non-sparse methods and those obtained using the sparse methods where specification of the regularization parameter automatically means model selection. In five out of six data sets, Partial Least Squares selects very dense networks. Furthermore, for data that violate the assumption of uncorrelated observations (due to replications), the Lasso and the adaptive Lasso yield very complex structures, indicating that they might not be suited under these conditions. The shrinkage approach is more stable than the regression based approaches when using subsampling.

Full text

Its licence does not allow us to show the full text. Read it at PubMed Central:

References 56

1.
Friedman N Inferring Cellular Networks using Probabilistic Graphical ModelsScience2004 3035659799805
Not in our library10.1126/science.1094068
2.
Yeung MKS Tegnér J Collins JJ Reverse Engineering Gene Networks using Singular Value Decomposition and Robust RegressionProceedings of the National Academy of Sciences2002 99961636168
Not in our library10.1073/pnas.092576199
3.
Rangel C Angus J Ghahramani Z Lioumi M Sotheran E Gaiba A Wild D Falciani F Modeling T-cell Activation using Gene Expression Profiling and State-Space ModelsBioinformatics2004 2013611372
Not in our library10.1093/bioinformatics/bth093
4.
Whittaker J Graphical Models in Applied Multivariate Statistics1990 Wiley New York
Not in our library
5.
Dobra A Hans C Jones B Nevins J Yao G West M Sparse Graphical Models for Exploring Gene Expression DataJournal of Multivariate Analysis2004 90196212
Not in our library10.1016/j.jmva.2004.02.009
6.
Schäfer J Strimmer K A Shrinkage Approach to Large-Scale Covariance Matrix Estimation and Implications for Functional GenomicsStatistical Applications in Genetics and Molecular Biology2005 432
Not in our library10.2202/1544-6115.1175
7.
Schäfer J Strimmer K An Empirical Bayes Approach to Inferring Large-Scale Gene Association NetworksBioinformatics2005 21754764
Not in our library10.1093/bioinformatics/bti062
8.
Li H Gui J Gradient Directed Regularization for Sparse Gaussian Concentration Graphs, with Applications to Inference of Genetic NetworksBiostatistics2008 72302317
Not in our library10.1093/biostatistics/kxj008
9.
Yuan M Lin Y Model Selection and Estimation in the Gaussian Graphical ModelBiometrika2007 941935
Not in our library10.1093/biomet/asm018
10.
Pihur V Datta S Datta S Reconstruction of Genetic Association Networks from Microarray DataBioinformatics2008 244561568
Not in our library10.1093/bioinformatics/btm640
11.
Ma S Gong Q Bohnert HJ An Arabidopsis Gene Network Based on the Graphical Gaussian ModelGenome Research2007 1716141625
Not in our library10.1101/gr.6911207
12.
Schmitt JE Lenroot RK Wallace GL Ordaz S Taylor KN Kabani N Greenstein D Lerch JP Kendler KS Neale MC Giedd JN Identification of Genetically Mediated Cortical Networks: A Multivariate Study of Pediatric Twins and SiblingsCerebral Cortex2008 18817371747
Not in our library10.1093/cercor/bhm211
13.
Efron B Large-Scale Simultaneous Hypothesis Testing: the Choice of a Null HypothesisJournal of the American Statistical Association2004 9996104
Not in our library10.1198/016214504000000089
14.
A unified approach to false discovery rate estimation · 2008
No notice
15.
Breiman L Bagging predictorsMachine Learning1996 24123140
Not in our library
16.
Tyekucheva S Chiaromonte F Augmenting the Bootstrap to Analyze High Dimensional Genomic DataTEST2008 17118
Not in our library10.1007/s11749-008-0098-6
17.
Strimmer K Comments on: Augmenting the Bootstrap to Analyze High Dimensional Genomic DataTEST2008 172527
Not in our library10.1007/s11749-008-0101-2
18.
Schäfer J Comments on: Augmenting the Bootstrap to Analyze High Dimensional Genomic DataTEST2008 172830
Not in our library10.1007/s11749-008-0102-1
19.
d'Aspremont A Banerjee O Ghaoui LE First-Order Methods for Sparse Covariance SelectionSIAM Journal on Matrix Analysis and its Applications2008 305666
Not in our library10.1137/060670985
20.
Rothman A Bickel P Levina E Zhu J Sparse Permutation Invariant Covariance EstimationElectronic Journal of Statistics2008 2494515
Not in our library10.1214/08-ejs176
21.
Witten D Tibshirani R Covariance-regularized regression and and classification for high-dimensional problemsJournal of Royal Statistical Society, Series B2009 713615636
Not in our library10.1111/j.1467-9868.2009.00699.x
22.
Yuan M Efficient Computation of ℓ1 Regularized Estimates in Gaussian Graphical ModelsJournal of Computational and Graphical Statistics2008 174809826
Not in our library10.1198/106186008x382692
23.
Tenenhaus A Guillemot V Gidrol X Frouin V Gene Association Networks from Microarray Data using a Regularized Estimation of Partial Correlation based on PLS RegressionIEEE Transactions on Computational Biology and Bioinformatics2008 http://doi.ieeecomputersociety.org/10.1109/TCBB.2008.87
Not in our library10.1109/tcbb.2008.87
24.
Wold H HMB et alPath Models with Latent Variables: The NIPALS ApproachQuantitative Sociology: International Perspectives on Mathematical and Statistical Model Building1975 Academic Press307357
Not in our library
25.
Wold S Ruhe H Wold H Dunn WJ III The Collinearity Problem in Linear Regression. The Partial Least Squares (PLS) Approach to Generalized InversesSIAM Journal of Scientific and Statistical Computations1984 5735743
Not in our library10.1137/0905052
26.
Saigo H Krämer N Tsuda K Partial Least Squares Regression for Graph Mining14th International Conference on Knowledge Discovery and Data Mining (KDD2008)2008 578586
Not in our library
27.
Boulesteix AL Strimmer K Partial Least Squares: a Versatile Tool for the Analysis of High-Dimensional Genomic DataBriefings in Bioinformatics2007 83244
Not in our library10.1093/bib/bbl016
28.
Rosipal R Trejo L Kernel Partial Least Squares Regression in Reproducing Kernel Hilbert SpacesJournal of Machine Learning Research2001 297123
Not in our library
29.
Rosipal R Krämer N Overview and Recent Advances in Partial Least SquaresSubspace, Latent Structure and Feature Selection Techniques, Lecture Notes in Computer Science2006 Springer3451
Not in our library
30.
Krämer N Braun ML Ghahramani ZKernelizing PLS, Degrees of Freedom, and Efficient Model SelectionProceedings of the 24th International Conference on Machine Learning2007 441448
Not in our library
31.
Sparse Partial Least Squares Regression for Simultaneous Dimension Reduction and Variable Selection · 2010
No notice
32.
Hoerl A Kennard R Ridge Regression: Biased Estimation for Nonorthogonal ProblemsTechnometrics2000 428086
Not in our library10.2307/1271436
33.
Meinshausen N Bühlmann P High Dimensional Graphs and Variable Selection with the LassoAnnals of Statistics2006 34314361462
Not in our library10.1214/009053606000000281
34.
Tibshirani R Regression Shrinkage and Selection via the LassoJournal of the Royal Statistical Society, Series B1996 58267288
Not in our library
35.
Zhou S Geer S van de Bühlmann P Adaptive Lasso for High Dimensional Regression and Gaussian Graphical Modeling2009 in press arXiv:0903.2515v1
Not in our library
36.
Zou H The Adaptive Lasso and its Oracle PropertiesJournal of the American Statistical Association2006 10147614181429
Not in our library10.1198/016214506000000735
37.
Krämer N Schäfer J parcor: estimation of partial correlations based on regularized regression2009 [R package version 0.1]
Not in our library
38.
Schäfer J Opgen-Rhein R Strimmer K Reverse Engineering Genetic Networks using the GeneNet PackageR News2006 5/65053
Not in our library
39.
Ruschhaupt M Erzeugung von positiv definiten Matrizen mit Nebenbedingungen zur Validierung von Netzwerkalgorithmen für Microarray-DatenPhD thesis2008 University of Munich
Not in our library
40.
Hastie T Efron B lars: Least Angle Regression, Lasso and Forward Stagewise2007 [R package version 0.9-7]
Not in our library
41.
Krämer N On the Peaking Phenomenon in Model Selection for the Lasso2009 in press http://arxiv.org/abs/0904.4416
Not in our library
42.
Kao K Yang Y Boscolo R Sabatti C Roychowdhury V Liao J Transcriptome-based Determination of Multiple Transcription Regulator Activities in Escherichia Coli by Using Network Component AnalysisProceedings of the National Academy of Sciences2004 1012641646
Not in our library10.1073/pnas.0305287101
43.
Schmidt-Heck W Guthke R Toepfer S Reischer H Duerrschmid K Bayer K Reverse engineering of the stress response during expression of a recombinant proteinEUNITE 2004 European Symposium on Intelligent Technologies, Hybrid Systems and their Implementation on Smart Adaptive Systems2004 407441
Not in our library
44.
Smith S Fulton D Chia T Thorneycroft D Chapple A Dunstan H Hylton C Zeeman S Smith A Diurnal Changes in the Transcriptome Encoding Enzymes of Starch Metabolism Provide Evidence for Both Transcriptional and Posttranscriptional Regulation of Starch Metabolism in Arabidopsis Leaves 1Plant Physiology2004 13626872699
Not in our library10.1104/pp.104.044347
45.
West M Blanchette C Dressman H Huang E Ishida S Spang R Zuzan H Olson J Jr Marks J Nevins J Predicting the Clinical Status of Human Breast Cancer by using Gene Expression ProfilesProceedings of the National Academy of Sciences2001 9821146211467
Not in our library10.1073/pnas.201162998
46.
Boulesteix AL Slawski M Stability and aggregation of ranked gene listsBriefings in Bioinformatics2009 10555668
Not in our library10.1093/bib/bbp034
47.
Saeys Y Inza I Larranaga P A review of feature selection techniques in bioinformaticsBioinformatics2007 23192507
Not in our library10.1093/bioinformatics/btm344
48.
Scutari M Structure variability in Bayesian networks2009 in press http://arxiv.org/abs/0909.1685
Not in our library
49.
Fleiss J Measuring nominal scale agreement among many ratersPsychological Bulletin1971 765378382
Not in our library10.1037/h0031619
50.
Boulesteix AL Kondylis A Krämer N Comment on: Augmenting the bootstrap to analyze high dimensional genomic dataTEST2008 173135
Not in our library10.1007/s11749-008-0103-0
51.
Reverter A Chan E Combining Partial Correlation and an Information Theory Approach to the Reversed-engineering of Gene Co-expression NetworksBioinformatics2008 242124912497
Not in our library10.1093/bioinformatics/btn482
52.
Pellet JP Elisseeff A Berthold MR, Shawe-Taylor J, Lavrac NA Partial Correlation-Based Algorithm for Causal Structure Discovery with Continuous VariablesAdvances in Intelligent Data Analysis VII, 7th International Symposium on Intelligent Data Analysis2007 229239
Not in our library
53.
Arnold A Liu Y Abe N Temporal Causal Modeling with Graphical Granger MethodsProceedings of the Thirteenth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining2007 ACM6675
Not in our library
54.
Strimmer K fdrtool: a versatile R package for estimating local and tail area-based false discovery ratesBioinformatics2008 2414611462
Not in our library10.1093/bioinformatics/btn209
55.
Boulesteix AL Lambert-Lacroix S Peyre J Strimmer K plsgenomics: PLS analyses for genomics2007 [R package version 1.2-2]
Not in our library
56.
Opgen-Rhein R Strimmer K longitudinal: Analysis of Multiple Time Course Data2008 [R package version 1.1.4]
Not in our library

Cited by papers we hold 4

Meta-Analysis of the Alzheimer’s Disease Human Brain Transcriptome and Functional Dissection in Mouse Models · 2020
No notice
Precision Lasso: accounting for correlations and linear dependencies in high-dimensional genomic data · 2018
No notice
Network analysis: a brief overview and tutorial · 2018
No notice
Estimating psychological networks and their accuracy: A tutorial paper · 2017
No notice

Reads

Read by people on this page

None yet.

Checked or read by AI through naulon

None yet.

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.

Are you an author?

Is this your paper?

Claim it to add your note and be paid when agents cite it. Free.