Why does the variate $min_long_intron_length set as 100000 in remove_long_intron_readthru_transcripts.pl

[Caizhengwen123]

Dears
I am confused why the $min_long_intron_length default value was 100000, is there any theoretical foundation support this setting?
For I turn on the '--human_gencode_filter' parameter, a mass of transcript was removed and caused false positive fusions, but if I turn off this parameter, identification of IGH and IGL fusions will almost be zero.
Hope for your reply!
Thanks!
Zhengwen CAI

[brianjohnhaas]

Hi, This setting is important for being able to detect fusions that involve long-range readthru transcripts or fusions due to chromosomal region deletions.

…

On Wed, Oct 28, 2020 at 2:46 AM ZhengWen Cai ***@***.***> wrote: Dears I am confused why the $min_long_intron_length default value was 100000, is there any theoretical foundation support this setting? For I turn on the '--human_gencode_filter' parameter, a mass of transcript was removed and caused false positive fusions, but if I turn off this parameter, identification of IGH and IGL fusions will almost be zero. Hope for your reply! Thanks! Zhengwen CAI — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#2>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABZRKXYUVABU32ZOQJLEWUTSM645DANCNFSM4TB4R76Q> .

-- -- Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas <http://broad.mit.edu/~bhaas>

[Caizhengwen123]

But when I use this parameter, much of genes' "possible readthrough transcripts" were deleted from the gtf annotation file, for example, the 'DDAH1-202' transcript was deleted, and then STAR-Fusion identified DDAH1 as a gene fusion with its neighbouring lncRNA 'AC092807', cause the 'DDAH1-202' have overlapping region with AC092807, below is the chromsome map of them:
http://asia.ensembl.org/Homo_sapiens/Gene/Summary?db=core;g=ENSG00000153904;r=1:85305965-85585847