I am trying to create an SQN file to submit an annotation to NCBI. When uploading the file I get missing transcript_ID and protein_ID. I double checked the examples from table2asn, and everything is as it should be. Does anyone have an example of what it should be like or an example of what it should look like in the SQN file? I have contacted ncbi multiple times but no response.
Thank you so much! Heleen
1 answer
Nine times out of ten this is the gnl| prefix. transcript_id and protein_id have to be namespaced as gnl|dbname|id - a bare value like mrna_00001 gets silently dropped, which is exactly the "missing" you're getting. So on the mRNA feature:
transcript_id gnl|YourCenter|mrna.LOCUSTAG_00001
protein_id gnl|YourCenter|LOCUSTAG_00001
and the CDS needs both qualifiers as well, with the same values as its mRNA. Leaving them off the CDS is the other common version of this.
They also have to be unique across the whole submission, and they're only required when mRNA features are present - a CDS-only prokaryotic table doesn't need them at all.
Log in to answer this question.
Example files are posted here: https://ftp.ncbi.nlm.nih.gov/asn1-converters/by_program/table2asn/DOCUMENTATION/
Are these what you were referring to above?
Thank you for the link, those files look very helpful! I will take a look and update if they helped me fix the issue.