Skip to main content
Fig. 1 | Algorithms for Molecular Biology

Fig. 1

From: Implications of non-uniqueness in phylogenetic deconvolution of bulk DNA samples of tumors

Fig. 1

Overview of the Perfect Phylogeny Mixture (PPM) problem. By comparing the aligned reads obtained from bulk DNA sequencing data of a matched normal sample and m tumor samples, we identify n somatic mutations and their frequencies \(F = \left[f_{p,c}\right]\). In the PPM problem, we are asked to factorize F into a mixture matrix U and a complete perfect phylogeny matrix B, explaining the composition of the m tumor samples and the evolutionary history of the n mutations present in the tumor, respectively. Typically, an input frequency matrix admits multiple distinct solutions. Here, matrix F has three solutions: (U, B), \((U',B')\) and \((U'',B'')\), where only (U, B) is the correct solution

Back to article page