Exon coordinates

[Egervari Lab]

I'm wondering if all coordinates provided in rMATS output are 0-based. Using rMATS v4.3.0 and specifically looking at {event type}.MATS.JCEC.txt files. 

I see _0base is sometimes specified (e.g., longExonStart_0base in A3SS & A5SS; 1stExonStart_0base & 2ndExonStart_0base in MXE; riExonStart_0base in RI; exonStart_0base in SE) but isn't always. 

Are the exon coordinates also 0-based when the _0base flag isn't included in the header? Or are those exon coordinates 1-based? 

Thanks in advance. 

[kutsc...@gmail.com]

The start coordinates are always 0-based while the end coordinates are always 1-based: https://github.com/Xinglab/rmats-turbo/issues/487#issuecomment-2815681610

Eric