The article introduces GENIE3, a new algorithm for inferring genetic regulatory networks (GRNs) from gene expression data. GENIE3 decomposes the problem into $p$ regression problems, each predicting the expression of one gene from the others using tree-based ensemble methods like Random Forests or Extra-Trees. The importance of each input gene in predicting the target gene's expression is used to indicate a potential regulatory link. These links are aggregated to reconstruct the entire network. GENIE3 performed well in the DREAM4 In Silico Multifactorial challenge and was also effective in deciphering the GRN of *Escherichia coli*. The algorithm is flexible, scalable, and can handle non-linear and combinatorial interactions, making it a promising tool for GRN inference.The article introduces GENIE3, a new algorithm for inferring genetic regulatory networks (GRNs) from gene expression data. GENIE3 decomposes the problem into $p$ regression problems, each predicting the expression of one gene from the others using tree-based ensemble methods like Random Forests or Extra-Trees. The importance of each input gene in predicting the target gene's expression is used to indicate a potential regulatory link. These links are aggregated to reconstruct the entire network. GENIE3 performed well in the DREAM4 In Silico Multifactorial challenge and was also effective in deciphering the GRN of *Escherichia coli*. The algorithm is flexible, scalable, and can handle non-linear and combinatorial interactions, making it a promising tool for GRN inference.