Hi,
I'm working on RNA-Seq data. Here's my basic workflow :
STAR -> cufflinks -> cuffmerge
So after that, I've a merged transcriptome gtf file. Now I want to annotate it to furthyer use it in differential expression analysis. So here's my idea. First extract the transcritp sequence using the gtf coordinates. Then, blast each sequence against ensembl transcript to annotate them.
What do you think ? Is there maybe another tool ?
thanks,
N.
•
written
by
Nicolas Rosewick
Convert cufflinks GTF into BED12 format using the following script and overlap the genomic-c0ordinates with known reference genome (RefSeq/UCSC/ENSEMBLE) by using BED tools. It gives you proper annotation. Here no need to to do BLAST and you can save lot of time.
For new transcripts that are not in the above databases, you need to add them. In your case, you can obtain all the gencode and human body map discovered lncRNAs (human-hg19) and add them to the reference genome BED file. That should work.
The problem is that the annotation of the species is not very good and a lot of things are missing (linke lncRNAs)
Log in to answer this question.
Hi, I am in similar situation. I have draft genome but no annotations. I need to do differential splicing analysis between conditions. How should I go about this ? Should I do tophat2 -> cufflinks -> cuffmerge -> Annotate GTF ?
If yes, then what would be the best way to annotate the GTF generated by cufflinks ?
Is there any alternative way for this ?