size difference when reciprocal minigraphing

[cistarsa]

Hello and thank you for developing this useful program. I'm attempting to identify SVs within a lineage for an introgressed trait between known markers. The size between these markers varies depending on the parental line, with the progeny exhibiting near perfect synteny with one of the parents, but not the other (expected). However, when I map all three individuals, as well as reciprocal graphs of offspring x parent1 and parent2 I get different graph sizes without obvious SVs, depending on the reference. For example the progeny against one parent produces a syntenic alignment with a total length of 1MB, whereas the reciprocal produces a syntenic graph of 3MB, each with one, s1, segment. What accounts for the ~2MB difference? Thank you.