Granger causality vs. dynamic Bayesian network inference: a comparative study.
Bottom Line:
For synthesized data, a critical point of the data length is found: the dynamic Bayesian network outperforms the Granger causality approach when the data length is short, and vice versa.We then test our results in experimental data of short length which is a common scenario in current biological experiments: it is again confirmed that the dynamic Bayesian network works better.When the data size is short, the dynamic Bayesian network inference performs better than the Granger causality approach; otherwise the Granger causality approach is better.
Affiliation: Department of Computer Science, University of Warwick, Coventry, UK. csrcbh@dcs.warwick.ac.uk
ABSTRACT
Show MeSH
Background: In computational biology, one often faces the problem of deriving the causal relationship among different elements such as genes, proteins, metabolites, neurons and so on, based upon multi-dimensional temporal data. Currently, there are two common approaches used to explore the network structure among elements. One is the Granger causality approach, and the other is the dynamic Bayesian network inference approach. Both have at least a few thousand publications reported in the literature. A key issue is to choose which approach is used to tackle the data, in particular when they give rise to contradictory results. Results: In this paper, we provide an answer by focusing on a systematic and computationally intensive comparison between the two approaches on both synthesized and experimental data. For synthesized data, a critical point of the data length is found: the dynamic Bayesian network outperforms the Granger causality approach when the data length is short, and vice versa. We then test our results in experimental data of short length which is a common scenario in current biological experiments: it is again confirmed that the dynamic Bayesian network works better. Conclusion: When the data size is short, the dynamic Bayesian network inference performs better than the Granger causality approach; otherwise the Granger causality approach is better. Related in: MedlinePlus |
Related In:
Results -
Collection
License getmorefigures.php?uid=PMC2691740&req=5
Mentions: In the next step, we extend our non-linear model to a more general setting in which the coefficients in the equations are randomly generated. Figure 6Aa shows the comparison result of the percentage of true positive connections derived from these two methods. It is very interesting to see that a critical point around 500 exists in the non-linear model, similar to the linear model before. From Figure 6Ab, the computing time required for the Bayesian network inference is still much larger than the Granger causality. In Figure 6B, we compare the performances on different coefficients (strength of interaction) for a fixed sample size of 900. From the five graphs, we can see that in general the Granger approach is more sensitive to a small value of the coefficients (see Figure 6B. X5 -> X4 and X4 -> X5). |
View Article: PubMed Central - HTML - PubMed
Affiliation: Department of Computer Science, University of Warwick, Coventry, UK. csrcbh@dcs.warwick.ac.uk
Background: In computational biology, one often faces the problem of deriving the causal relationship among different elements such as genes, proteins, metabolites, neurons and so on, based upon multi-dimensional temporal data. Currently, there are two common approaches used to explore the network structure among elements. One is the Granger causality approach, and the other is the dynamic Bayesian network inference approach. Both have at least a few thousand publications reported in the literature. A key issue is to choose which approach is used to tackle the data, in particular when they give rise to contradictory results.
Results: In this paper, we provide an answer by focusing on a systematic and computationally intensive comparison between the two approaches on both synthesized and experimental data. For synthesized data, a critical point of the data length is found: the dynamic Bayesian network outperforms the Granger causality approach when the data length is short, and vice versa. We then test our results in experimental data of short length which is a common scenario in current biological experiments: it is again confirmed that the dynamic Bayesian network works better.
Conclusion: When the data size is short, the dynamic Bayesian network inference performs better than the Granger causality approach; otherwise the Granger causality approach is better.