Inconsistency between jaffa_results.csv and intermediate files

[Puputnik]

Hi!

I'm using JAFFAL with ONT data, i'm writing because i'm interested in getting the sequences of the spanning reads supporting a gene fusion (or the IDs, so i can extract them from the input fastq).
I thought could use the intermediate files ".fusions.fa" or ".reads", but as far as i've understood they contain unfiltered reads (for a certain fusion gene i see less supporting reads in jaffa_results.csv than in these files); also, not all the fusions reported in jaffa_results.csv are present in the intermediate files (is this expected? this sounds strange to me, but probably i'm missing a point. where do these calls come from since their supporting reads are not reported at all in intermediate files?).

Can you please help me shedding light on this? i'm out of ideas
Thanks a looot in advance for the help!!

All the best