More posts like this
-
whole genome analyses
written by Rima •
-
how to assembly the RNASeq data
written by xinhui.wang
-
Downloading RNA-Seq Raw Count data form ICGC
written by David_emir
-
smallRNA aligner and pipeline
written by noeD
-
TopHat2: --read-realign-edit-dist = 0 or default options?
written by noeD
-
variant calling tumor/normal exome sequencing
written by noeD
-
high read depth and threshold for filtering
written by noeD
-
DNA-seq Breast Cancer dataset
written by pingEde •
-
amplicon longer than read
written by pingEde •
-
How to remove primers from a Roche 454 sff file
written by pingEde •
(Modern, mid-2016) RNA-seq software pipeline
this: http://www.nature.com/nprot/journal/v11/n9/full/nprot.2016.095.html
Even though I would use Hisat2, Cufflinks and DESeq2 for alignment, quantification and differential expression analysis.
What analyses do you need? RNASeq data analysis can be a broad category!
thank you for your reply :)
I have several RNA-seq data from different samples (so I think that I should also normalize the ouputs among samples) and I want to do differential expression analysis. I would need help especially for the first phase of analysis and with normalization.
Thank you!
DESeq2 and edgeR are two R packages that do most of that for you. As @Macspider mentioned above your "pipeline" would need to get you a matrix of "raw" read counts. This is the required input DE packages mentioned at the beginning of this post. Lots more details about DESeq2 are in the vignette.
Thank you for your reply. Probably I really didn't understand right how DESeq2 works... because I am looking for a tool that allow me to normalize my data across samples in order to make the expression comparable.
Thanks
So you are only looking to do normalization but not interested in doing differential expression analysis as a subsequent step?
I apologise for the confusion of my replay, I am also interested in it, but I am wondering how to deal the normalization across sample :)
DESeq2 can do that for you. See this thread for how to get those: which normalized counts is better from DESeq2
That said DESeq2 uses integer data counts as default input.
If you feed DESeq2 normalized results, you'll have worse results :) It wants raw counts and will do the job for you. Be careful to set the "design" variable according to your experimental design. There are many tutorials on how to set it depending on how your experiment is constructed!