Invalid cluster_id for abundance script about cdna_cupcake

[Magdoll]

Hi,
Please try the latest Cupcake version that i just pushed in. It should recognize the Iso-Seq3 ID formats now. If not, let me know and I can help make sure the formats work.

--Liz

[SziKayLeung]

Hello Liz,

Sorry for the late reply but thank you so much for updating the scripts. The formats worked, and the abundance.txt and group.stat.txt outputs are as expected :)

However, during the run, I received loads of warnings "WARNING: isoseq3 format detected. Output length column will be NA." (specifically, 179,600 of these lines, with no other information).

I don't recall this happening with Isoseq2(beta) on smrtlink. So just wanted to check with you if this is something that I should be expecting with Isoseq3, and what are reads that are giving NA length.

Thank you,
Szi Kay

[Magdoll]

Hi @SziKayLeung ,

That warning was intentionally added in the last Cupcake update. This is because older versions of the FL CCS id format allowed extraction of sequence lengths. The ID format in Iso-Seq1 and 2 used to be <movie>/<zmw>/<start>_<end>_CCS but now in Iso-Seq 3 it is <movie>/<zmw>/ccs and the length information is not contained in the ID. It's a minor matter but to kept compatibility I'm still outputting the length field, but you'll find them NA when running on Iso-Seq 3 runs.

--Liz

[SziKayLeung]

Thank you @Magdoll for clarifying!