Failure to call well-known fusions on WES samples

[maximus3219]

I have performed WES on multiple dermatofibrosarcoma protuberans (DFSP) samples with known COL1A1::PDGFB fusion.
However manta fails to call this translocation on all the samples.
I have changed the two values 'minEdgeObservations' and 'minCandidateSpanningCount' in the manta configuration file to 2.
And supply --callRegions with the targeted regions.

Is there any way to further increase the sensitivity?
Also, is there anyway I can annotate the translocation with the gene names? Occasional samples have call of translocation involving t(17;22), but the genomic location seems to be falling out of COL1A1 and PDGFB gene, respectively. Not sure if they are the correct calls.....