The identification of fusion genes

[qiu527]

Hi Brian,

I have a question about the diagram. As shown in the figure, if the transcribed RNA includes a portion from the forward-transcribed gene A and another portion from the reverse-transcribed gene B, would this fusion still be reported as a fusion between gene A and gene B?

Thank you very much for your time and assistance.

Best regards,
Marina

[brianjohnhaas]

Hi Marina, It won't call fusions in that situation. Only the fusions that are transcribed in a sense::sense orientation will be reported. (Note, if they're antisense::antisense, then STAR-Fusion should flip it around and report it as sense::sense ). hope this helps, Brian

…

On Mon, Jun 9, 2025 at 10:24 AM Marina ***@***.***> wrote: *qiu527* created an issue (STAR-Fusion/STAR-Fusion#429) <#429> Hi Brian, I have a question about the diagram. As shown in the figure, if the transcribed RNA includes a portion from the forward-transcribed gene A and another portion from the reverse-transcribed gene B, would this fusion still be reported as a fusion between gene A and gene B? fusion.png (view on web) <https://github.com/user-attachments/assets/9055eac9-12e7-4f1b-b4d5-2c89d57f9f9f> Thank you very much for your time and assistance. Best regards, Marina — Reply to this email directly, view it on GitHub <#429>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABZRKX3M3XB2OZJGX4SZPU33CWKKHAVCNFSM6AAAAAB65CMJ52VHI2DSMVQWIX3LMV43ASLTON2WKOZTGEZTANBXGU3TIOA> . You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

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

[sufyazi]

Hi Brian,

If the first part is sense shouldn't it theoretically be transcribing A-into-B anyway despite the antisense orientation of the 3' partner? even if there is a premature stop codon, that shouldn't matter right and such "aberrant" form of fusion protein can still be interesting to look at especially in the context of diseases?

Are you saying STAR-Fusion only explicitly returns sense::sense fusion events by default? If yes, is there a way for end user to skip this? Should I not look at the abridged output then?

[brianjohnhaas]

I don't think STAR-fusion will report any fusions that aren't in the sense::sense orientation. It only focuses on fusion events that involve parts of the normally transcribed regions of known genes. It doesn't do a more general structural variation type of call based on chimeric transcript alignments. So there could be plenty of evidence for transcriptional units that support a structural rearrangement, but STAR-Fusion will restrict itself to those events that generate the types of transcriptional products that are known to be most relevant to cancer - which almost entirely involve sense::sense orientations to my knowledge.

hope this helps