Figure 1

Workflow of VERSE. (a) Reads are aligned to a host reference genome. Unmapped reads and read pairs with one end unmapped are called viral reads. To differentiate mapped reads in the figure from unmapped ones, which remain grey throughout the pipeline, the color of mapped reads is changed from grey to the color of the host genomic region they are aligned to. (b) The viral reads are mapped to a virus reference genome. The high-quality consensus SNPs and indels detected from aligned reads are used to modify the virus reference genome. (c) The consensus virus genome created is concatenated to the host reference genome (designated as a separate pseudo-chromosome, chrVirus). Next, the viral reads are mapped to the resulting new reference. Then, inter-chromosomal structural variants (SVs) are detected from aligned reads. The SVs involving both the host genome and chrVirus are used to infer virus integration-harboring regions in the host genome. Finally, using the same procedure as in (b), the identified host genomic regions are customized. (d) The modified host genomic regions are concatenated with the consensus virus genome. The viral reads are mapped to this new reference for the detection of inter-chromosomal SVs. The breakpoints of the SVs that involve both the virus and host genomes, if there are any, are reported as virus integration sites. In the figure, vertical dotted lines represent virus integration breakpoints.