Improving long-read somatic structural variant calling with pangenome and de novo personal genome assembly

Here is a more formal, Zenodo-style version: These datasets provide supplementary resources supporting the minisv manuscript and include four main categories of data. First, the collection contains somatic structural variant (SV) calling results generated by Severus, nanomonsv, and SAVANA in somatic calling mode. Calls were produced using the GRCh38 reference before and after de novo assembly-based filtering, based on paired tumor–normal long-read whole-genome sequencing data from COLO829, HCC1395, HCC1937, HCC1954, NCI-H1437, and NCI-H2009. Corresponding results from downsampled datasets are also included. Second, the collection includes mosaic SV calling results generated by Severus and Sniffles2 in mosaic calling mode. These calls were produced using the GRCh38 reference before and after de novo assembly-based filtering from mixed tumor–normal long-read whole-genome sequencing data for COLO829. Third, the datasets include de novo assembly FASTA files for the matched normal samples from cancer paired datasets, including COLO829BL, HCC1395BL, HCC1937BL, HCC1954BL, NCI-H1437BL, and NCI-H2009BL. Downsampled-data-based de novo assemblies are also provided.