Vân Anh Huynh-Thu
GENIE3Three implementations of GENIE3 are available:
You have also the possibility to run GENIE3 (MATLAB implementation) on the GP-DREAM platform.
GENIE3 is based on regression trees. To learn these trees, the Python implementation uses the scikit-learn library, the MATLAB implementation wraps a C code written by Pierre Geurts, and the R implementation uses the randomForest package. The running times of the different GENIE3 implementations are shown below for the DREAM5 networks (in each case, GENIE3 was run using the default parameters).
Inferring regulatory networks from expression data using tree-based methods
Huynh-Thu, V. A., Irrthum, A., Wehenkel, L., and Geurts, P.
PLoS ONE, 5(9):e12776, 2010.
Note: All the results presented in the paper were generated using the MATLAB implementation.
GENIE3 was the best performer in two DREAM challenges :
- the DREAM5 Network Inference challenge
- the DREAM4 In Silico Size 100 Multifactorial sub-challenge
GENIE3 variantsWe derived two extensions of GENIE3, for the inference of networks from time series data and systems genetics data (i.e. expression data + genomic data) respectively.
The GENIE3 extension for time series data is described in chapter 8 of my thesis.
The GENIE3 extension for systems genetics data is described in the following book chapter:
Gene regulatory network inference from systems genetics data using tree-based methods
Huynh-Thu V. A., Wehenkel L., and Geurts P.
In: A. de la Fuente (Ed.), Gene Network Inference - Verification of Methods for Systems Genetics Data, Springer, pp. 63-85, 2013.
Python and MATLAB implementations of both GENIE3 extensions are available. Please contact me if you are interested.
- MATLAB [Zip file]
Combining tree-based and dynamical systems for the inference of gene regulatory networks
Huynh-Thu, V. A. and Sanguinetti, G.
Bioinformatics, 31(10):1614-1622, 2015.