missed junction

[gorliver]

Hi Dan,
I found some juctions were misses in the gff output(See attached igv screenshot).

As you can see, the missed junction is well covered by junction reads. The alignments were generated using STAR and portcullis were run using default parameters.
The alignments were from cross-species mapping, so there are plenty of SNPs in exons and probably close the junction sites. Do you think the SNPs will cause problem or did I missed something?
Many thanks,
Gorliver