This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
A lag operator ℒ is such that ℒX t =X t−1. Thus, applying the operator k times yields: ℒk X t =X t−k . The auto-regressive model X t +A 1 X t−1+A 2 X t−2+⋅⋅⋅=ε t can be represented as (Id+A 1ℒ+A 2ℒ2+⋅⋅⋅)X t =ε t .
References
R. Aebersold, L.E. Hood, and J.D. Watts. Equipping scientists for the new biology. Nat Biotechnol, 18(4):359, 2000.
H. Akaike. A new look at the statistical model identification. Autom Control, IEEE Trans, 19(6):716–723, 1974.
U. Alon. Biological networks: the tinkerer as an engineer. Science, 301(5641):1866–1867, 2003.
D. Anastassiou. Computational analysis of the synergy among multiple interacting genes. Mol Syst Biol, 3:83, 2007.
C. Andrieu, N. de Freitas, A. Doucet, and M.I. Jordan. An introduction to MCMC for machine learning. Mach Learn, V50(1):5–43, 2003.
W.L. Buntine. Operations for learning with graphical models. J Artif Intell Res, 2:159, 1994.
D.M. Camacho and J.J. Collins. Systems biology strikes gold. Cell, 137(1):24–26, 2009.
I. Cantone, L. Marucci, F. Iorio, M.A. Ricci, V. Belcastro, M. Bansal, S. Santini, M. di Bernardo, D. di Bernardo, and M.P. Cosma. A yeast synthetic network for in vivo assessment of reverse-engineering and modeling approaches. Cell, 137(1):172–181, 2009.
B. Chance, R.W. Estabrook, and A. Ghosh. Damped sinusoidal oscillations of cytoplasmic reduced pyridine nucleotide in yeast cells. Proc Natl Acad Sci, 51(6):1244–1251, 1964.
Y. Chen, S.L. Bressler, and M. Ding. Frequency decomposition of conditional Granger causality and application to multivariate neural field potential data. J Neurosci Methods, 150(2):228–237, 2006.
J.J. Chrobak and G. Buzsaki. Gamma oscillations in the entorhinal cortex of the freely behaving rat. J Neurosci, 18:388–398, 1998.
M. Ding, Y. Chen, and S.L. Bressler. Granger causality: Basic theory and application to neuroscience. In J. Timmer, B. Schelter, M. Winterhalder, editors, Handbook of Time Series Analysis, pages 451–474. Wiley-VCH, Weinheins, 2006.
M.R. Doyle, S.J. Davis, R.M. Bastow, H.G. McWatters, L. Kozma-Bognár, F. Nagy, A.J. Milla, and R.M. Amasino. The elf4 gene controls circadian rhythms and flowering time in arabidopsis thaliana. Nature, 1419:74–77, 2002.
T. Fawcett. An introduction to ROC analysis. Pattern Recogn Lett, 27(8):861–874, 2006.
J.F. Feng, D.Y. Yi, R. Krishna, S.X. Guo, and V. Buchanan-Wollaston. Listen to genes: dealing with microarray data in the frequency domain. PLoS ONE, 4(4):e5098+, 2009.
K. Friston. Causal modelling and brain connectivity in functional magnetic resonance imaging. PLoS Biol, 7(2):e1000033+, 2009.
K.J. Friston, L. Harrison, and W. Penny. Dynamic causal modelling. NeuroImage, 19(4):1273–1302, 2003.
T.S. Gardner, D. di Bernardo, D. Lorenz, and J.J. Collins. Inferring genetic networks and identifying compound mode of action via expression profiling. Science, 301(5629):102–105, 2003.
T. Ge, K.M. Kendrick, and J.F. Feng. A unified dynamic and granger causal model approach demonstrates brain hemispheric differences during face recognition learning. PLoS Comput Biol, 2009, submitted.
J.F. Geweke. Measurement of linear dependence and feedback between multiple time series. J Am Stat Assoc, 77(378):304–313, 1982.
J.F. Geweke. Measures of conditional linear-dependence and feedback between time series. J Am Stat Assoc, 79(388):907–915, 1984.
B. Gourévitch, R.L. Bouquin-Jeannès, and G. Faucon. Linear nonlinear causality between signals: methods, examples and neurophysiological applications. Biol Cybern, 95(4):349–369, 2006.
C. Granger. Investigating causal relations by econometric models and cross-spectral methods. Econometrica, 37:424–438, 1969.
C. Granger. Testing for causality: a personal viewpoint. J Econ Dynam Control, 2:329–352, 1980.
S. Guo, A.K. Seth, K.M. Kendrick, C. Zhou, and J.F. Feng. Partial Granger causality–eliminating exogenous inputs and latent variables. J Neurosci Methods, 172(1):79, 2008.
S. Guo, J. Wu, M. Ding, and J.F. Feng. Uncovering interactions in the frequency domain. PLoS Comput Biol, 4(5):e1000087, 2008.
S. He. Estimation of the mixed AR and hidden periodic model. Acta Math Appl Sin Engl Ser, 13(2):196–208, 1997.
E. Klipp, R. Herwig, A. Kowald, C. Wierling, and H. Lehrach. Systems biology in practice: concepts, implementation and application, 2005.
C. Ladroue, S.X. Guo, K. Kendrick, and J.F. Feng. Beyond element-wise interactions: identifying complex interactions in biological processes. PLoS ONE, 4(9):e6899, 2009.
J.C. Locke, L. Kozma-Bognar, P.D. Gould, B. Feher, E. Kevei, F. Nagy, M.S. Turner, A. Hall, and A.J. Millar. Experimental validation of a predicted feedback loop in the multi-oscillator clock of arabidopsis thaliana. Mol Syst Biol, 2:59, 2006.
H.G. McWatters, E. Kolmos, A. Hall, M.R. Doyle, R.M. Amasino, P. Gyula, F. Nagy, A.J. Millar, and S.J. Davis. ELF4 is required for oscillatory properties of the circadian clock. Plant Physiol, 144(1):391, 2007.
D.S. Moore. The Basic Practice of Statistics. Freeman, New York, 2003.
M. Morf, A. Vieira, D.T.L. Lee, and T. Kailath. Recursive multichannel maximum entropy spectral estimation. Geosci Electron IEEE Trans, 16(2):85–94, 1978.
S. Mukherjee and T.P. Speed. Network inference using informative priors. Proc Natl Acad Sci, 105(38):14313–14318, 2008.
C.J. Needham, J.R. Bradford, A.J. Bulpitt, and D.R. Westhead. A primer on learning in Bayesian networks for computational biology. PLoS Comput Biol, 3(8):e129, 2007.
A. Neumaier and T. Schneider. Estimation of parameters and eigenmodes of multivariate autoregressive models. ACM Trans Math Softw, 27(1):27–57, 2001.
J. Pearl. Causality: Models, Reasoning, and Inference. Cambridge University Press, Cambridge, 2000.
J. Quackenbush. Computational analysis of microarray data. Nat Rev Genet, 2(6):418–427, 2001.
K. Sachs, O. Perez, D. Pe’er, D.A. Lauffenburger, and G.P. Nolan. Causal protein-signaling networks derived from multiparameter single-cell data. Science, 308(5721):523–529, 2005.
M. Schelter, B. an Winterhalderm, and J. Timmer. Handbook of Time Series Analysis: Recent Theoretical Developments and Applications. Wiley-VCH, Weinheim, 2006.
T.F. Schultz and S.A. Kay. Circadian clocks in daily and seasonal control of development. Science, 301(5631):326–328, 2003.
T.P. Speed. Statistical Analysis of Gene Expression Microarray Data. CRC Press, Boca Raton, 2003.
A.N. Stepanova and J.M. Alonso. Arabidopsis ethylene signaling pathway. Science, 276:1872–1874, 2005.
G.C. Tiao and M.R. Grupe. Hidden periodic autoregressive-moving average models in time series data. Biometrika, 67(2):365–373, 1980.
H.R. Ueda. Systems biology flowering in the plant clock field. Mol Syst Biol, 2:60, 2006.
H.R. Ueda, W.B. Chen, A. Adachi, H. Wakamatsu, S. Hayashi, T. Takasugi, M. Nagano, K. Nakahama, Y. Suzuki, S. Sugano, M. Iino, Y. Shigeyoshi, and S. Hashimoto. A transcription factor response element for gene expression during circadian night. Nature, 418(6897):534–539, 2002.
N. Wiener. The theory of prediction. Mod Math Eng Ser, 1:125–139, 1956.
J.H. Wu, K. Kendrick, and J.F. Feng. Detecting correlation changes in electrophysiological data. J Neurosci Methods, 161(1):155–165, 2007.
J.H. Wu, X.G. Liu, and J.F. Feng. Detecting causality between different frequencies. J Neurosci Methods, 167(2):367–375, 2008.
J.H. Wu, J.L. Sinfield, and J.F. Feng. Impact of environmental inputs on reverse-engineering approach to network structures. BMC Systems Biology, 3:113, 2009.
J. Yu, A.V. Smith, P.P. Wang, and A.J. Hartemink. Advances to Bayesian network inference for generating causal networks from observational biological data. Bioinformatics, 20(18):3594–3603, 2004.
C.L. Zou and J.F. Feng. Granger causality vs. dynamic bayesian network inference: a comparative study. BMC Bioinform, 10(1):122, 2009.
C.L. Zou, C. Ladroue, S.X. Guo, and J.F. Feng. Identifying interactions in the time and frequency domains in local and global networks. BMC Bioinform, 2010, under revision.
C.L. Zou, K.M. Kendrick, and J.F. Feng. The fourth way: Granger causality is better than the three other reverse-engineering approaches. Cell, 2009. http://www.cell.com/comments/S0092-8674(09)00156-1.
M. Zylka, L. Shearman, J. Levine, X. Jin, D. Weaver, and S. Reppert. Molecular analysis of mammalian timeless. Neuron, 21(5):1115–1122, 1998.
Acknowledgements
This work is supported by grants from EPRSC (UK, CARMEN EP/E002331/1) and EU grant (BION) and NSFC (China).
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix: Estimating the Error Covariance Matrix
Appendix: Estimating the Error Covariance Matrix
The main quantity of interest for calculating Granger causality is the covariance matrix of the error term ε in the d-dimensional auto-regressive (AR) model
The coefficients of the matrices A i need to be estimated from the data. Typically, they are found by minimising the variance of the error between the prediction X t and the observation at the same time t. Morf’s procedure [33] provides a fast and robust way of estimating A i and the covariance matrix of ε via a recursive algorithm. Unless prior knowledge informs us about the likely order p of the AR model (for example by knowing at which time-scale one should expect interactions to take place), it also has to be estimated from the data. The goodness of fit alone is not sufficient for selecting the optimal order: adding a new order implies adding d 2 more unknowns to the system, which considerably improves the fitting simply by increasing the degrees of freedom. The Akaike Information Criterion (AIC, [2]) provides a trade-off between the model complexity (a function of p) and the fit. For an AR model of dimension d, of order p and with T observations, this quantity can be written as
The optimal order p is then defined as the one that minimises the AIC. A point estimate of the Granger causality is often not sufficient for making a strong conclusion about the data and a confidence interval is required. In some cases ([25] for partial Granger causality), it is possible to derive the confidence interval in closed form. In general, it is estimated via a bootstrap procedure [32]—given the optimal AR model, an ensemble of signals are generated which each produce a value for the Granger causality. The final estimation of the Granger causality and its confidence interval are defined as the average and the standard error of this set of estimates.
Rights and permissions
Copyright information
© 2010 Springer-Verlag London Limited
About this chapter
Cite this chapter
Guo, S., Ladroue, C., Feng, J. (2010). Granger Causality: Theory and Applications. In: Feng, J., Fu, W., Sun, F. (eds) Frontiers in Computational and Systems Biology. Computational Biology, vol 15. Springer, London. https://doi.org/10.1007/978-1-84996-196-7_5
Download citation
DOI: https://doi.org/10.1007/978-1-84996-196-7_5
Publisher Name: Springer, London
Print ISBN: 978-1-84996-195-0
Online ISBN: 978-1-84996-196-7
eBook Packages: Computer ScienceComputer Science (R0)