During cancer development, the tumor cell population usually emerges from a single cell ancestor and can therefore be categorized as clonal. As a result of the genomic instability and selection within this clonal population, additional mutations take place and allow for the differentiation of cell lineages. This tumor heterogeneity can be captured by doing clonality analysis which stratifies distinct tumor cell populations into groups, called subclones. The process of estimating the subclonal composition of a tumor requires intricate algorithmic approaches. Due to the growing popularity of clonality analysis, many tools have been made available for the reconstruction of subclonal architectures. The majority of the tools use next generation sequencing data to infer aspects of the subclonal composition and its dynamics. The vastness of available tools calls for an unbiased comparative method between them. Here a novel framework is presented to satisfy this requirement. By integrating a selection of available tools into this framework and testing their response to different simulated data, some of the tool qualities can be identified. The achieved results show this method is able to compare the PhyloWGS and DPClust tools using in silico generated tumor samples. Next to this, the framework is capable of analyzing real data. Samples taken from metastatic sites of 8 bladder cancer patients will be discussed.