Question: Best Program to use for DE Analysis
0
gravatar for oma219
18 months ago by
oma21930
oma21930 wrote:

Hello,

I was wondering if anyone had advice on the best program to use for DE Analysis after using STAR for alignment? I'm not sure whether to use CuffLinks or DESeq. Also, do both of them accept .bam files that are produced by STAR. Thanks.

Omar

rna-seq • 730 views
ADD COMMENTlink modified 18 months ago by Biojl1.6k • written 18 months ago by oma21930
1

Most people here would advice to use DESeq2, or similar tools such as limma trend/voom or edgeR.

You'll have to get raw read counts first from your bam files, I prefer using featureCounts for this.

ADD REPLYlink written 18 months ago by b.nota4.9k

Is Cufflinks also a possible option for DE analysis or is it used more in other situations? Thanks!

ADD REPLYlink written 18 months ago by oma21930
2

Cufflinks is no longer advised software for RNA-seq analysis, an updated pipeline is described here.

ADD REPLYlink written 18 months ago by WouterDeCoster32k

Thanks! So would you agree that DESeq2 is one of the better programs to use for DE Analysis? I just finished the alignment with STAR so I was trying to decide which programs would be best.

ADD REPLYlink written 18 months ago by oma21930
1

DESeq2 is indeed a very good tool to do gene expression analysis. You can go for it. The R vignette is well done and you can find a lot of tutorial by googling it.

ADD REPLYlink written 18 months ago by Nicolas Rosewick6.6k
2
gravatar for Biojl
18 months ago by
Biojl1.6k
Barcelona
Biojl1.6k wrote:

Actually I would give a try to Sleuth, the DE tool based on pseudo-alignments. Is by far the most user friendly, accurate and easy to use. It has other advantages, take a look: http://pachterlab.github.io/sleuth/

You will have to run kallisto or salmon to get the quantification (it does not rely on traditional alignments.). These programs are super fast and can be even run on a laptop in minutes.

ADD COMMENTlink written 18 months ago by Biojl1.6k
4

It's worth noting that Sleuth does have it's drawbacks if you have a more complex experimental design than A vs B, at least it did when I last used it, and that's not to say that I know of any other solid alternatives. Additionally, you can use Salmon to quantify (similar concept to Kallisto), and put it through Wasabi to make it ready for Sleuth

ADD REPLYlink written 18 months ago by andrew.j.skelton735.3k
3

Is by far the most user friendly, accurate and easy to use.

Compared to other common DE tools (Cufflinks, DESeq2, edgeR), Sleuth is definitely not the most user friendly or easy.

ADD REPLYlink written 18 months ago by igor6.6k
Please log in to add an answer.

Help
Access

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
Powered by Biostar version 2.3.0
Traffic: 1050 users visited in the last hour