In this article, we present the re_actShap method that allows the detection of rewiring events in the Gene Regulatory Networks (GRNs). The novelty of this method is that the only required inputs are a sc-RNA-Seq datasets dataset, and a list of transcription factors (TFs) for the species. We propose to train machine learning models to predict the expression of target genes with the expression of the TFs. Then, we use an explainability method based on the SHAP values: topShap. With the output of topShap, we build an association matrix, in which each cell is represented by its activation of the GRNs regulatory links. We use a Student test to detect the regulatory associations that are differentially activated in each cell type. We use several visualizations to generate hypotheses on the rewiring events in the GRN. The method is compatible with the tool scanpy, widely used to analyze sc-RNA-Seq datasets.