R packages for gene regulatory networks from time series data

Question

[Ann]

I am looking for software (preferably an R package) that can accept a matrix of gene expression values from a time course experiment and then compute a possible gene regulatory network based on the data.

[Ann]

Also I'm looking at aracne2, Califano lab. But would like to try several options.

Answer 1

[pengchy]

I am looking for the same answers. The following is the results of my literature review:

First, three review papers should be checked:

• Lopes, M. and G. Bontempi (2013). "Experimental assessment of static and dynamic algorithms for gene regulation inference from time series expression data." Front Genet 4: 303.
• Kim, Y., et al. (2014). "Inference of dynamic networks using time-course data." Brief Bioinform 15(2): 212-228.
• Liu, Z. P. (2015). "Reverse Engineering of Genome-wide Gene Regulatory Networks from Gene Expression Data." Curr Genomics 16(1): 3-22.

According to the review paper, several packages are available currently for construct transcription regulatory network from time series expression data:

• GRNInfer, Huber group LASSO, these two can construct the regulatory network from multiple time series expression data.
• TDARACNE, DPC.
• Bayes network: Banjo, BNFinder, GlobalMIT
• TESLA
• KELLER
• ARTIVA

Hope this will help you. Any suggestions on this topic are welcome!

Answer 2

[Renesh]

Try this: http://labs.genetics.ucla.edu/horvath/CoexpressionNetwork/Rpackages/WGCNA/

[Ann]

Thanks! Any others that infer causal relationships from time series data? (Horvath package seems more general, focused on co-expression networks.)

[Renesh]

For this case, you need to perform correlation analysis and use of any test statistics (eg. z test) to identify significant co regulation. For this case, you need large sample size (> 10) for reliable result.

Answer 3

[raffael.azevedo]

Hi, did you tried RTN package?

Answer 4

[joram]

ARACNE is pretty good