Closed diekhans closed 3 years ago
This is the case of both transcriptomes uploaded by Cindy.
The ONT one had pretty similar IDs so I could fix it easily. For example, the isoform with a transcript_id "0037c387-bdb2-4952-a592-8e475de7e5bc" in the GTF file was called in the expression matrix as "0037c387-bdb2-4952-a592-8e475de7e5bc_chr7:28560000", which is actually the concatenation of the given transcript_id and gene_id.
However, for the transcriptome generated from PacBio data transcript_ids looked like CCS ids (e.g. "m54284U_191111_172039/57606429/ccs"), but the IDs reported in the expression matrix do not match at all (00000f84-64f2-4aef-a027-a86468e48da2;16). Probably that was an error uploading the wrong file...
ids are now validated
from mock submissions
One submission with IDs that do not match with the reported ones in the GTF file → Transcripts can be evaluated, but not their expression