I need to include new genes, pseudogenes and modify gene/intron-exon boundaries of few genes in the available annotations from NCBI. GFF3 files follow complex format which is very difficult to modify. I need a tool which can make the task easier without hampering downstream analysis like finding differentially expressed genes, etc. Is there any other way of doing this that I am missing?
There are a number of tools that allow you to modifiy annotations and export the changes to a file with gff/gtf format.
One of the multiple options is the NCBI Genome Workbench, which is free, (http://www.ncbi.nlm.nih.gov/tools/gbench/)
Best,
Juan
maybe checkout this list also:
https://en.wikipedia.org/wiki/List_of_gene_prediction_software
although most of them will be old.
also you can use Trinity / rsem as tools (complete workflow).
links:https://github.com/trinityrnaseq/trinityrnaseq/wiki and https://github.com/deweylab/RSEM
for handling the GTF format maybe look around for tools that can do this like the tools on usegalaxy.org or finding/using the sanger gtf tools or the brentlab gtf tool linked on http://mblab.wustl.edu/GTF22.html#resources
Do you have additional evidence to support your modifications?
There is a section about merging/resolve legacy annotations with Maker (see link below)
http://weatherby.genetics.utah.edu/MAKER/wiki/index.php/MAKER_Tutorial_for_GMOD_Online_Training_2014#Merge.2FResolve_Legacy_Annotations
I have homology-based information for a protein family, but many of those need to be manually edited due to faulty intron-exon boundaries, missing fragments incorporated by automated techniques. I need a smart tool which can help me do that easily. I tried to edit GTF file manually but it lead to complications later - hence the question.
Thank you very much for your inputs!
Snehal
Hello Julianne,
At that time, I struggled with the installation of Apollo, but then gave up after a while. I should have ideally contacted the developers then with the errors I was encountering. I was dealing with ~150 sequences for which I had a good semi-automated way to get accurate internal boundaries. I just needed refinement of the 5' and 3' ends, which I decided to do manually 😅. I made a bed file, which was a bit simpler than GFF file that could be handled by STAR, DEseq etc..
However, since then I have come across more genome annotation editors (not just viewers) like Artemis, Ugene in addition to Apollo that might help in the process in a better way. I would definitely spend much more time exploring them, if I am to do a similar project again.
Hope this helps you.
Snehal
When I say I manually edited a bed file, I mean I looked at the predicted gene boundaries in a genome viewer, in my case simple NCBI's viewer, and changed the frame I am looking at to be close to the ends of the gene to find typical START and STOP codons. I noted these down separately in the bed file. This is not the best way to do it, and I would definitely avoid it this time around. 😁
I'll just throw out Apollo: http://github.com/GMOD/Apollo/. We put together a docker build of it. We are mostly focused around GFF, but support the visualization of all types via Jbrowse (https://github.com/GMOD/JBrowse). It's an active and funded project (for the next few years), so I would definitely give it a shot and ask questions if you get stuck.
Thank you for sharing information of docker builds @Nathan. @Julianne hope this helps you.
- Badges
- Science method