Pvclust is an R package designed to assess the uncertainty in hierarchical cluster analysis, particularly useful for applications such as DNA microarray analysis. It employs bootstrap resampling techniques to calculate probability values (p-values) for each cluster, specifically the approximately unbiased (AU) p-value and the bootstrap probability (BP) value. The package supports multiscale bootstrap resampling, which is more biased than the ordinary bootstrap resampling but offers better computational efficiency. Pvclust can be easily installed and used in R, with parallel computing options available for faster processing. The package includes functions for graphical representation and detailed information about p-values. An example using lung tumor gene expression data demonstrates the effectiveness of Pvclust in assessing cluster reliability. The authors recommend using a default number of bootstrap replicates for initial testing and adjusting the number for smaller errors. The package is freely available under the GNU General Public License and can be installed from CRAN.Pvclust is an R package designed to assess the uncertainty in hierarchical cluster analysis, particularly useful for applications such as DNA microarray analysis. It employs bootstrap resampling techniques to calculate probability values (p-values) for each cluster, specifically the approximately unbiased (AU) p-value and the bootstrap probability (BP) value. The package supports multiscale bootstrap resampling, which is more biased than the ordinary bootstrap resampling but offers better computational efficiency. Pvclust can be easily installed and used in R, with parallel computing options available for faster processing. The package includes functions for graphical representation and detailed information about p-values. An example using lung tumor gene expression data demonstrates the effectiveness of Pvclust in assessing cluster reliability. The authors recommend using a default number of bootstrap replicates for initial testing and adjusting the number for smaller errors. The package is freely available under the GNU General Public License and can be installed from CRAN.