Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks

Marco Grzegorczyk; Dirk Husmeier

doi:10.1007/978-3-642-04031-3_11

Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks

Marco Grzegorczyk^*, Dirk Husmeier

^*Corresponding author for this work

School of Biological Sciences

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Feedback loops and recurrent, structures are essential to the regulation and stable control of complex biological systems. The application of dynamic as opposed to static Bayesian networks is promising in that, in principle, these feedback loops can be learned. However, we show that the widely applied BGe score is susceptible to learning spurious feedback loops, which are a consequence of non-linear regulation and autocorrelation in the data. We propose a non-linear generalisation of the BGe model, based on a mixture model, and demonstrate that this approach successfully represses spurious feedback loops.

Original language	English
Title of host publication	PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS
Editors	Kadirkamanathan, G Sanguinetti, M Girolami, M Niranjan, J Noirel
Place of Publication	BERLIN
Publisher	Springer
Pages	113-124
Number of pages	12
ISBN (Print)	978-3-642-04030-6
DOIs	https://doi.org/10.1007/978-3-642-04031-3_11
Publication status	Published - 2009
Event	4th International Conference Pattern Recognition in Bioinformatics - Sheffield, United Kingdom Duration: 7 Sept 2009 → 9 Sept 2009

Publication series

Name	LECTURE NOTES IN BIOINFORMATICS
Publisher	SPRINGER-VERLAG BERLIN
Volume	5780
ISSN (Print)	0302-9743

Conference

Conference	4th International Conference Pattern Recognition in Bioinformatics
Country/Territory	United Kingdom
Period	7/09/09 → 9/09/09

Keywords / Materials (for Non-textual outputs)

ALLOCATION SAMPLER
MODEL

Access to Document

10.1007/978-3-642-04031-3_11

SynthSys; formerly CSBE: Centre for Systems Biology at Edinburgh
Millar, A., Beggs, J., Ghazal, P., Goryanin, I., Hillston, J., Plotkin, G., Tollervey, D., Walton, A. & Robertson, K.
BBSRC
8/01/07 → 31/12/12
Project: Research

Cite this

Grzegorczyk, M., & Husmeier, D. (2009). Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks. In Kadirkamanathan, G. Sanguinetti, M. Girolami, M. Niranjan, & J. Noirel (Eds.), PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS (pp. 113-124). (LECTURE NOTES IN BIOINFORMATICS; Vol. 5780). Springer. https://doi.org/10.1007/978-3-642-04031-3_11

@inproceedings{16d5dd6be19644859ab7bb3f688ffda1,

title = "Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks",

abstract = "Feedback loops and recurrent, structures are essential to the regulation and stable control of complex biological systems. The application of dynamic as opposed to static Bayesian networks is promising in that, in principle, these feedback loops can be learned. However, we show that the widely applied BGe score is susceptible to learning spurious feedback loops, which are a consequence of non-linear regulation and autocorrelation in the data. We propose a non-linear generalisation of the BGe model, based on a mixture model, and demonstrate that this approach successfully represses spurious feedback loops.",

keywords = "ALLOCATION SAMPLER, MODEL",

author = "Marco Grzegorczyk and Dirk Husmeier",

year = "2009",

doi = "10.1007/978-3-642-04031-3_11",

language = "English",

isbn = "978-3-642-04030-6",

series = "LECTURE NOTES IN BIOINFORMATICS",

publisher = "Springer",

pages = "113--124",

editor = "Kadirkamanathan and G Sanguinetti and M Girolami and M Niranjan and J Noirel",

booktitle = "PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS",

address = "United Kingdom",

note = "4th International Conference Pattern Recognition in Bioinformatics ; Conference date: 07-09-2009 Through 09-09-2009",

}

Grzegorczyk, M & Husmeier, D 2009, Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks. in Kadirkamanathan, G Sanguinetti, M Girolami, M Niranjan & J Noirel (eds), PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS. LECTURE NOTES IN BIOINFORMATICS, vol. 5780, Springer, BERLIN, pp. 113-124, 4th International Conference Pattern Recognition in Bioinformatics, United Kingdom, 7/09/09. https://doi.org/10.1007/978-3-642-04031-3_11

Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks. / Grzegorczyk, Marco; Husmeier, Dirk.
PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS. ed. / Kadirkamanathan; G Sanguinetti; M Girolami; M Niranjan; J Noirel. BERLIN: Springer, 2009. p. 113-124 (LECTURE NOTES IN BIOINFORMATICS; Vol. 5780).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks

AU - Grzegorczyk, Marco

AU - Husmeier, Dirk

PY - 2009

Y1 - 2009

N2 - Feedback loops and recurrent, structures are essential to the regulation and stable control of complex biological systems. The application of dynamic as opposed to static Bayesian networks is promising in that, in principle, these feedback loops can be learned. However, we show that the widely applied BGe score is susceptible to learning spurious feedback loops, which are a consequence of non-linear regulation and autocorrelation in the data. We propose a non-linear generalisation of the BGe model, based on a mixture model, and demonstrate that this approach successfully represses spurious feedback loops.

AB - Feedback loops and recurrent, structures are essential to the regulation and stable control of complex biological systems. The application of dynamic as opposed to static Bayesian networks is promising in that, in principle, these feedback loops can be learned. However, we show that the widely applied BGe score is susceptible to learning spurious feedback loops, which are a consequence of non-linear regulation and autocorrelation in the data. We propose a non-linear generalisation of the BGe model, based on a mixture model, and demonstrate that this approach successfully represses spurious feedback loops.

KW - ALLOCATION SAMPLER

KW - MODEL

U2 - 10.1007/978-3-642-04031-3_11

DO - 10.1007/978-3-642-04031-3_11

M3 - Conference contribution

SN - 978-3-642-04030-6

T3 - LECTURE NOTES IN BIOINFORMATICS

SP - 113

EP - 124

BT - PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS

A2 - Kadirkamanathan, null

A2 - Sanguinetti, G

A2 - Girolami, M

A2 - Niranjan, M

A2 - Noirel, J

PB - Springer

CY - BERLIN

T2 - 4th International Conference Pattern Recognition in Bioinformatics

Y2 - 7 September 2009 through 9 September 2009

ER -

Grzegorczyk M, Husmeier D. Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks. In Kadirkamanathan, Sanguinetti G, Girolami M, Niranjan M, Noirel J, editors, PATTERN RECOGNITION IN BIOINFORMATICS, PROCEEDINGS. BERLIN: Springer. 2009. p. 113-124. (LECTURE NOTES IN BIOINFORMATICS). doi: 10.1007/978-3-642-04031-3_11

Avoiding Spurious Feedback Loops in the Reconstruction of Gene Regulatory Networks with Dynamic Bayesian Networks

Abstract

Publication series

Conference

Keywords / Materials (for Non-textual outputs)

Access to Document

Fingerprint

Projects

SynthSys; formerly CSBE: Centre for Systems Biology at Edinburgh

Cite this