Return exon that is adjacent to a genomic breakpoint

[jarbesfeld]

This method will be used in FUSOR when a genomic breakpoint does not occur within an exon, and the closest exon to the breakpoint is needed to create a TranscriptSegmentElement.

[ahwagner]

Is "closest" the right metric here? Perhaps it should be "first adjacent", and be orientation specific? Consider the following:

Here is a rearrangement (in + strand orientation) that results in a Gene A::Gene B fusion, with Gene A in a position that is 5' of Gene B. The red break occurs closest to Gene A Exon 2, but the first 5' adjacent is Gene A Exon 1. Similarly, the orange break is closest to Gene B Exon 4, but the first 3' adjacent is Gene B Exon 5.

[jarbesfeld]

@ahwagner Do you have a similar diagram for the - strand orientation?

[ahwagner]

Yes. Here is an example where Gene A is on the reverse strand. The genomic segment with this gene is broken and attached inverted to the normal reference orientation–indicated by the break asymmetry (red line lower in the rearrangement graphic).

The exon selection should be identical to the first example.