Algorithms to Model Single Gene, Single Chromosome, and Whole Genome Copy Number Changes Jointly in Tumor Phylogenetics

摘要: We present methods to construct phylogenetic models of tumor progression at the cellular level that include copy number changes scale single genes, entire chromosomes, and whole genome. The are designed for data collected by fluorescence in situ hybridization (FISH), an experimental technique especially well suited characterizing intratumor heterogeneity using counts probes genetic regions frequently gained or lost development. Here, we develop new provably optimal computing edit distance between states two cells given evolution probes, all on a chromosome, then apply this theory practical heuristic algorithm, implemented publicly available software, inferring phylogenies from potentially hundreds evolutionary model. demonstrate validate simulated published FISH cervical cancers breast cancers. Our computational experiments show model algorithm lead more parsimonious trees than prior single-tumor phylogenetics improved performance various classification tasks, such as distinguishing primary tumors metastases obtained same patient population.