I'm using BreakDancer Max (v 1.1.2) to identify structural variants in multiple dog samples (WGS of blood cells) that have cancer. Does anyone have suggestions for optimal filtering? Currently as a threshold for selecting deletions, I'm using 1) a confidence score greater than or equal to 90 and 2) the median of total supporting reads across the entire sample output. Wondering if there's another approach people have used in their past analyses.