vg sv calling

Question

[twesigomwedavid]

Hello,

Does vg have a way of handling imprecise SV breakpoints? I have been trying to genotype a deletion test case but the breakpoints differ in different samples (e.g. in homozygotes vs heterozygotes). Thanks

Answer 1

[glenn.hickey]

No, unfortunately not. Nodes and edges in vg's graph model correspond to exact positions. In theory, you can store every possible deletion that you see in your data within one graph. But in practice, there is a point of diminishing returns when adding too much similar variation to a graph, and it's often best to select a representative variant.