Wrong gene classification about cdna_cupcake

[Trandamere]

I suggest you separate the positive and negative chains to classify.

[Magdoll]

@Trandamere ,

This may be a bug. The different strands are processed independently so this should not happen. Would it be possible to share the mapped reads (fasta or fastq) and SAM file for just this region? If it's mapping to a known ref like hg19 or hg38, I will only need the read fasta/fastq. You can email it to [email protected].

[Trandamere]

@Magdoll
This is the fasta and SAM file of PB583
PB583.zip

[Magdoll]

Hi @Trandamere ,
Thanks for sharing the dataset! You are right...while the strands were processed separately, in the initial grouping stage they were grouped together. I've pushed the code fixes to Cupcake version 5.2 (if you use tags, it is tagged tofu2_v19)

You should now see three gene groups (PB.1 and PB.2 for the two + strand genes, PB.3 for the - strand).

Please let me know if this fixes your problems.

--Liz

[Trandamere]

@Magdoll
Hi,The proble is is solved after using Cupcake version 5.2!