hi everyone i have assembled genome sequence, assembled transcriptome sequence and two sample of transcripts reads and i mapped these reads to transcriptome with BWA. i want to know how to calculate RPKM value after mapping.is there any software or tool which calculate RPKM value? please tell me. and second thing is that when i used cuffdiff it wants ref.gtf file and i don't have ref.gtf file. from where i can get it. please reply as soon as possible thank you.
•
written
by
raj.gzra •
Since August I stopped using cuffdiff because some of the results produced by the latest version look quite improbable. I switched to using the following pipeline:
There is a very nice tutorial ( http://www-huber.embl.de/pub/pdf/nprot.2013.099.pdf) - you just need to adapt the scripts from it and you will have a working pipeline. The pipeline won't directly compute the RPKM values but I am told that edgeR has a function (rpkm()) which would do this for you. Instead of using that function I used R script to compute the RPKM values which I picked from this post (http://seqanswers.com/forums/showpost.php?p=120485&postcount=5). By the way, RPKM is still a very popular metric but it has some downsides (http://www.ncbi.nlm.nih.gov/pubmed/22872506). You would be better off using the FDR and logFC computed by edgeR (or DESeq).
You would need a gtf file that corresponds to the genome version that you have used to align the reads. The best solution is to download them both from Ensemble (http://www.ensembl.org/info/data/ftp/index.html).
thank you nikolay. but i have genome of fungus which is unannotated yet and newly sequenced and assembled. then how to download files from Ensemble>>
can i make gtf file by blast against genome with transcriptome???
Log in to answer this question.
