SpliceAI aggregation fails

[PandaRescuer]

Dear @ReverendCasy

I was excited to see the release of TOGA2 2.0.7. Thank you for maintaining the project.

I encountered an error when running toga2.py spliceai using the human T2T-CHM13v2.0 genome.

The run fails during SpliceAI aggregation with the following error:
[shared.py] - CRITICAL: File aggregation for AcceptorPlus failed:: There's more than one value for NC_060930.1 base 12000001 (in coordinates that start with 1).

I tried reducing the chunk size to 1,000,000, but the same error occurred again, this time at a different chromosome and position:
NC_060929.1 base 162000001

Could you please advise how this problem should be resolved?

Thank you again for providing such an innovative and powerful toolkit. I really appreciate all the work your team has put into TOGA and TOGA2.