Abstract
Tumourigenesis is a mutation accumulation process, which is likely to start with a mutated founder cell. The evolutionary nature of tumor development makes phylogenetic models suitable for inferring tumor evolution through genetic variation data. Copy number variation (CNV) is the major genetic marker of the genome with more genes, disease loci, and functional elements involved. Fluorescence in situ hybridization (FISH) accurately measures multiple gene copy number of hundreds of single cells. We propose an improved binary differential evolution algorithm, BDEP, to infer tumor phylogenetic tree based on FISH platform. The topology analysis of tumor progression tree shows that the pathway of tumor subcell expansion varies greatly during different stages of tumor formation. And the classification experiment shows that tree-based features are better than data-based features in distinguishing tumor. The constructed phylogenetic trees have great performance in characterizing tumor development process, which outperforms other similar algorithms.
Highlights
Cancer is the most serious and dangerous disease to human health in the world
Tumor phylogenetic tree inference can be treated as minimum Steiner tree problem in directed graph, which is a NP-hard problem
Two Fluorescence in situ hybridization (FISH) datasets, cervical cancer and breast cancer, respectively, from Wangsa et al [43] and HeselmeyerHaddad et al [44], are published to visualize copy number changes in tumors based on single-cell analyses
Summary
Over the past few decades, researchers have been working on the diagnosis and treatment of cancer. Owing to these great efforts, our understanding of cancer has been greatly improved, and early clinical diagnosis and reliable treatment are critical for cancer [1]. Cancer is the result of an imbalance in the cell cycle of the organism. Each cell of the organism contains a complete genome and has great spontaneity [1]. When the genome is no longer regulated by normal tissue and the spontaneity of cells is activated, cancer develops. Tumor cells succumb to different evolutionary pressures and result in constant replication, growth, invasion, and metastasis [1]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
