Quantification of a gene that is not in the reference genome

Question

[felipead6]

I work with mouse genome and I have RNA-Seq data from a mouse strain. When I align the reads in the mouse genome, a gene is not quantified since this gene does not appear in the reference mouse genome. How can I quantify this gene?

Answer 1

[Jake Warner]

Manually add it to the genome as its own contig and re-align!
Edit: To be a little more explicit, you would append your reference fasta with the gene sequence of interest using something like cat musmus.fa newgene.fa > mus_edited.fa then add an entry in the GFF file for the gene, then re-build the reference index and re-align your reads.

[Nicolas Rosewick]

You don't have to realign. Just add the gene of interest informations in the gtf file and rerun featurecounts (to count reads per gene) and then use edgeR or DESeq2.

[felipead6]

That sounds interesting. How do I add the gene information in the .gtf file considering that I have the position of the gene available?

[Nicolas Rosewick]

look at : http://www.ensembl.org/info/website/upload/gff.html for gtf column description, and add corresponding informations to the gtf (so open it in a text editor and edit it).

[Karma]

just open the gtf with gedit or excel and fill

[Nicolas Rosewick]

Do not use excel please : https://genomebiology.biomedcentral.com/articles/10.1186/s13059-016-1044-7 https://bmcbioinformatics.biomedcentral.com/articles/10.1186/1471-2105-5-80

[Jake Warner]

If the sequence is present and the annotation doesn't exist then this is indeed the correct approach!

[felipead6]

I have the .gb file from genbank, is there any way to transform this to .gtf?

[Joe]

That part is easily googled:

e.g:

https://github.com/riverlee/genbank2gtf