p-SCNAClonal: Somatic copy number alterations based tumor subclonal population inferring method

摘要

When using next generation sequencing (NGS) data of tumor and its paired normal to detect somatic copy number alteration (SCNA) regions, the imbalance read's PCR amplification process and system noise introduced by the twice sequencing procedures cause the result of existing algorithms to contain many false positive breakpoints and SCNA regions. These false positive breakpoints and SCNA regions further affect SCNA based subclonal population inferring tool that uses these SCNA regions as input. We present p-SCNAClonal, a tool that improves tumor subclonal population inferring by merging the SCNA regions according to the read count and B-allele frequency (BAF) information to reduce the number of false positive SCNA breakpoints. p-SCNAClonal then locates the baseline SCNA (not containing any SCNAs) and infers the absolute copy number of SCNA and its subclonal populations through a probabilistic model. We show p-SCNAClonal's superiority to existing SCNA based subclonal population inferring method. p-SCNAClonal is publicly available as a Python package at https://github.com/Billy-Nie/pSCNAClonal.