Istvan Albert I tried to use bio gff to convert a genebank file into a gff, it produced this error: TypeError: '>' not supported between instances of 'NoneType' and 'int'
The gene bank accession number is EF489041.1 Thanks!
Hello
Background I'm trying to analyse a RNA-seq experiment of Bacillus subtilis PY79. As part of that, I need to create an ensemble database using ensembldb (https://bioconductor.org/packages/release/bioc/html/ensembldb.html). For this I need a gff file of the genome. I tried to download it from NCBI, however, I get an error because that gff file lacks "gene_id". Since I cannot find any other gff file of that subspecies, I am now trying to generate it from the gb file.
The Problem I have a genebank (gb) file which I have downloaded from NCBI (https://www.ncbi.nlm.nih.gov/nuccore/NC_022898.1?report=genbank then send to -> file -> GeneBank (full)). I wish to convert it to a gff3 file. I have attempted several things, but no succeeded.
What I've Tried
This usecase is one of the many for which I wrote the bio package. Get the file:
bio fetch NC_022898
convert the Genbank file to GFF like so:
bio convert NC_022898 --gff > annotations.gff
See more here:
https://www.bioinfo.help/bio-gff.html
note how much nicer and prettier the gene models made with bio are (plus are fully compatible with featureCounts):
Disclaimer: the package is still under heavy development and has not been sufficiently tested
Istvan Albert I tried to use bio gff to convert a genebank file into a gff, it produced this error: TypeError: '>' not supported between instances of 'NoneType' and 'int'
The gene bank accession number is EF489041.1 Thanks!
open an issue or a discussion on the source code website so that this can be fixed ther
Log in to answer this question.
The GFF file for this strain does have
geneidentifier. You should be able to use that for your counting usingfeatureCounts. This is bacterial RNAseq so things are simpler. Align with aligner of your choice and then usefeatureCountswith-g geneoption. If you choose this file then be sure to get the corresponding genome fasta file to create your indexes. That way all identifiers will match.