Fig. 3

Visual comparison of the copy number profiles of LNCaP cancer cell line generated by OneD, HiNT and HiCNAtra. a Coverage plots showing genome-wide CNV profiles estimated by OneD (first panel), HiNT (second panel) and HiCNAtra (third panel) derived from LNCaP Hi-C data. The last panel shows the CNAtra estimated CNV profiles derived from WGS data of LNCaP. Note: The centromeres and telomeres of chromosomes are filtered out in CNAtra output. b Zoom-in view of the CNV profiles of chr 4 and 5 show that CNV3 and CNV4 are not detected by OneD and HiNT, respectively. Each grey dot represents the copy number of a bin. The black line represents the copy number track where any amplitude transition indicates a new CNV region. For HiCNAtra, the copy number track shown here is computed from the LCVs only. The red blocks indicate CNVs that are missed or incorrectly called by OneD or HiNT. c The CN frequency distribution exhibits the multimodality feature of the LNCaP cancer genome with most genomic segments having a copy number of 4. The vertical black line denotes the copy number reference (CN = 2)