Does STAR-Fusion identify new fusions?

[DadongZ]

I noticed some top identified fusions are not in the list of ./StarFv1.6/GRCh38_gencode_v29_CTAT_lib_Mar272019.plug-n-play/ctat_genome_lib_build_dir/trans.blast.align_coords.align_coords.dat
Does it indicate this is a new fusion event?

[brianjohnhaas]

Hi, Previously identified fusions will show up in the Fusion Annotator database that gets bundled into the source repo, and compiled into the database that travels with the plug-n-play distribution. These fusions get annotated when they're found among the results. The trans.blast.align_coords.align_coords.dat file is used by STAR-Fusion, but it doesn't contain fusion events. Instead it indicates the regions of sequence similarity found between annotated gene targets. hope this helps, ~b

On Wed, May 8, 2019 at 11:38 AM DadongZhang ***@***.***> wrote: I noticed some top identified fusions are not in the list of

./StarFv1.6/GRCh38_gencode_v29_CTAT_lib_Mar272019.plug-n-play/ctat_genome_lib_build_dir/trans.blast.align_coords.align_coords.dat

…

Does it indicate this is a new fusion event? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.

-- -- Brian J. Haas The Broad Institute http://broadinstitute.org/~bhaas

[DadongZ]

Thanks!
Where could I found the known fusion list that used in Star Fusion?

[brianjohnhaas]

here you go: https://github.com/FusionAnnotator/CTAT_HumanFusionLib/releases

…

On Wed, May 8, 2019 at 11:47 AM DadongZhang ***@***.***> wrote: Thanks! Where could I found the known fusion list that used in Star Fusion? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#133 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABZRKX4QPDJE4NLNWMIFQBTPULYY7ANCNFSM4HLS4EXQ> .

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

[DadongZ]

Thanks very much!

[DadongZ]

Is there a good way to check whether a specific gene is in the list (I saw a list of files and not sure which one is the best to check)?

[brianjohnhaas]

It uses the gencode annotation. In the CTAT genome lib, you can just look at the ref_annot.gtf file to see all the genes that are considered. If you need to use a different annotation, that's possible too, but it requires that you go through the CTAT genome lib building process. best, ~b

…

On Wed, May 8, 2019 at 11:58 AM DadongZhang ***@***.***> wrote: Is there a good way to check whether a specific gene is in the list (I saw a list of files and not sure which one is the best to check)? — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#133 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABZRKX4U4WYPBYZ5JYZZFCDPUL2BDANCNFSM4HLS4EXQ> .

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

[DadongZ]

Thanks! We are good at using GENCODE annotation. So if a gene in the ref_annot.gtf, Star fusion will predict its related fusion events? I am asking this is because we were told a gene should be very likely in a fusion but we didn't find it and wanted make sure the known fusions of the gene do in the annotation list.

[brianjohnhaas]

You can try to grep the gene name from the preliminary fusion list to see if anything came up with weak evidence. Also, try some other tools to see if they find it as well. FusionCatcher, Arriba, and ChimPipe are all very good too. best, ~b

…

On Wed, May 8, 2019 at 1:06 PM DadongZhang ***@***.***> wrote: Thanks! We are good at using GENCODE annotation. So if a gene in the ref_annot.gtf, Star fusion will predict its related fusion events? I am asking this is because we were told a gene should be very likely in a fusion but we didn't find it and wanted make sure the known fusions of the gene do in the annotation list. — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#133 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABZRKXZYQ5OWDXBR5VIFJHDPUMCC7ANCNFSM4HLS4EXQ> .

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

[DadongZ]

Thanks!