Dear all,
My question might be quit specific but hopefully there is someone that might have some advice for me. I am trying to upload a .sqn file to the NCBI genome database. I received the error that my sequence/gene headers are too long. The headers are automatically generated by the annotation software and ideally I do not want to change these in my original GFF and genome assembly files. But how do I change the headers in the .sqn only? I could simply create a list of long names that I rename into "gene1, gene2, gene3" etc.
But I never worked with a .sqn file so I would like to ask you if the above would be possible?
One of the error lines look like this:
ERROR: File: sexigua_aseembly_annot.sqn, Code(SEQ_INST_BadSeqIdFormat), Sequence-id: gnl|HF086|augustus_masked-tulip_contig_306_pilon_pilon-processed-gene-0.1-mRNA-1:cds, General identifier longer than 50 characters
Where "augustus_masked-tulip_contig_306_pilon_pilon-processed-gene-0.1-mRNA-1" should be replaced by something shorter.
Thanks in advance for your ideas and suggestions!
0 answers
No answers yet.
Log in to answer this question.
-
Concatenating text files based on common indices
written by Hau Tak Leighton •
-
to grep pattern
written by sherafzalk769 •
-
How to extract only the specific fasta sequences by matching the ID's to a different source files ?
written by pinn
-
htseq gtf problem
written by kti.adler •
-
how to change the sequence ID and order of sequences in a file to match an alignment output file
written by Kay •
-
Filtering GFF annotations file
written by alslonik
-
Making Zero based coordinate system in gff file
written by 1234anjalianjali1234 •
-
Splitting a fasta file based on the number of headers.
written by sicat.paolo20 •
-
Create a presence absence matrix in R for gene data
written by natasha
-
merge two files
written by biolab
If you do that would that not throw your entire dataset out of sync?
This sounds pretty non-informative/important. Can that not be replaced everywhere? What does
0.1refer to?On your first question, I was afraid of that too. But this .sqn file is the only file I upload so it contains the genome assembly (does not get changed) and the information of the GFF file. Because I do want to edit the gene headers in all occasions within this file I think it should be working out fine (see comment below). At least worth a trial.
Just to add, I have created a text file with in the first column the old gene name and in the second column the new gene name:
Ideally, with a sed like command each occurrence in the .sqn file of the first column gets replaced by the name in the second column. Not sure if thats how a .sqn file works but worth a trial. If anyone knows how to use sed in combination with a separate file to replace 1000s of strings than please let me know! I will post it here if I have found a solution too.
Even if the
.sqnfile gets through NCBI's initial check the ID's would then not match GTF/Sequence files. Are you replacing these ID's in all files?But what I understood from the NCBI tutorials (but it is quit hard to get everything!) is that this .sqn file is the only file I need to upload and is derived from the assembly and the GFF file. Am I right here? If I still need to upload the GFF file also, than indeed this will not work.
I actually now think to do the gene name replacement on the original GFF file. In that case I am sure that the resulting .sqn is built correctly and I have not accidentally changed anything in the .sqn that shouldn't be changed!
That sounds like a good plan.