Standard Processing Steps for RNASeq and Bisulfite-Seq raw data

[Fernas]

Hi all,

I have the raw mRNA-Seq/Bisulfite-Seq data for multiple cell types (in sra format). I want to know if there is a research article or website where I can find the standard pre-processing steps for both types of data. For example:
1) quality based filtering
2) read trimming
3) coverage threshold
4) mapping criteria (multiple mapping issue,...etc)
...etc

[fkrueger]

We are processing our internal data pretty much according to this protocol.

[Fernas]

Thank you very much indeed @fkrugeger. It is very useful documentation. I am wondering if there is something similar to this for RNA-Seq?

Regarding your documentation, I read it and I have couple of questions:

1) How can we cacluate the methylation proportion: for example, I found that some research articles calculated methylation proportions as (methylated calls / (methylated calls + unmethylated calls)). How can I calculate this?

2) Filtering for high read coverage: I am wondering if we have different libraries, shall we set the coverage threshold to be the same for all replicates/libraries? or it depends on the library size? what threshold you used?

3) Pool replicates: If I have 4 libraries represent 4 different technical or biological replicates of the same cell type. The document did not discuss how can I combine the replicates. Any idea about that?

4) Did you apply any normalization/re-scale methods to remove the sequencing bias between libraries and/or cell types?

[crazyhottommy]

http://www.nature.com/nprot/journal/v8/n9/full/nprot.2013.099.html

and

http://www.nature.com/nprot/journal/v7/n3/full/nprot.2012.016.html

[peromhc]

For read trimming in RNAseq, please see: http://journal.frontiersin.org/Journ...00013/abstract