Mfuzz is an R package for soft clustering of microarray data, offering a more flexible approach compared to traditional hard clustering methods. Soft clustering allows genes to belong to multiple clusters, improving the ability to detect complex patterns in gene expression data. The package includes a graphical user interface (Mfuzzgui) for ease of use. Mfuzz uses the fuzzy c-means algorithm to minimize variation within clusters, resulting in gradual membership values that reflect the degree of association between genes and clusters. This method is more robust to noise and can avoid the need for pre-processing steps that might lose important information. The software accepts input data in various formats, including tables and Bioconductor objects, and allows users to adjust parameters such as the number of clusters and fuzzification parameter. The output includes a partition matrix with membership values, which can be used to identify cluster cores and improve the detection of regulatory mechanisms. Results can be further processed within Bioconductor or exported in simple table format. Mfuzz is not limited to microarray data and has been applied to protein phosphorylation time series. The software is available under the GPL version 2 license and is freely available. It is an R package, requiring Tcl/Tk for the graphical interface, and includes scripts for installation and standalone use. Future developments include enhanced export options, such as HTML reports. The study was supported by a grant from the German Research Foundation. The authors thank colleagues for their assistance and feedback.Mfuzz is an R package for soft clustering of microarray data, offering a more flexible approach compared to traditional hard clustering methods. Soft clustering allows genes to belong to multiple clusters, improving the ability to detect complex patterns in gene expression data. The package includes a graphical user interface (Mfuzzgui) for ease of use. Mfuzz uses the fuzzy c-means algorithm to minimize variation within clusters, resulting in gradual membership values that reflect the degree of association between genes and clusters. This method is more robust to noise and can avoid the need for pre-processing steps that might lose important information. The software accepts input data in various formats, including tables and Bioconductor objects, and allows users to adjust parameters such as the number of clusters and fuzzification parameter. The output includes a partition matrix with membership values, which can be used to identify cluster cores and improve the detection of regulatory mechanisms. Results can be further processed within Bioconductor or exported in simple table format. Mfuzz is not limited to microarray data and has been applied to protein phosphorylation time series. The software is available under the GPL version 2 license and is freely available. It is an R package, requiring Tcl/Tk for the graphical interface, and includes scripts for installation and standalone use. Future developments include enhanced export options, such as HTML reports. The study was supported by a grant from the German Research Foundation. The authors thank colleagues for their assistance and feedback.