Matching read IDs for each transcript

[andyjslee]

I want to see which read IDs support each assembled transcript. Is the following the correct way to do this?

1. The transcript ID is the value corresponding to s description in rnabloom.transcripts.fa (e.g. s=1).
2. Look for the above value in the 6th column in the rnabloom.longreads.assembly3.map.paf.gz file.

If an assembled transcript description in the rnabloom.transcripts.fa file does not have a s= information but instead has path=[237857+,237856+], what is the transcript ID? Is it an ambiguous assembled transcript?

[kmnip]

The s and path indicate which seed sequences were used to generate that sequence.
For example, path=[1+,2+,3-] means the sequence was generated from a path through sequences 1, 2, and 3 from the set of sequences in the previous step. The + and - indicate the orientation of the sequence. Also, the IDs imply nothing about "ambiguous" transcripts. Since there are multiple steps in the assembly, the sequence IDs do not always correspond to the input read IDs.