due to very helpful suggestions from Fengshen (David) Kuo, I have revised the code for importing paired-end reads. It is not necessary to distinguish between bwa or bowtie in MEDIPS version >=1.23.2.
Please see https://www.bioconductor.org/developers/how-to/useDevel/
on instructions how to use packages in the development branch. The development package is also available at https://www.bioconductor.org/packages/3.4/bioc/html/MEDIPS.html
. Next release will be in ~ Nov 2016.
As you have three groups, I recommend to have a look into our novel package qsea, which comes with improved MeDIP specific normalisation and allows for more complex designs: https://www.bioconductor.org/packages/3.4/bioc/html/qsea.html
All the best,
On 08 Aug 2016, at 16:08, stb [bioc] <email@example.com<mailto:firstname.lastname@example.org>> wrote:
Activity on a post you are following on support.bioconductor.org<https: support.bioconductor.org=""/>
User stb<https: support.bioconductor.org="" u="" 11175=""/> wrote Question: MeDIPS and use of the 'bwa' parameter<https: support.bioconductor.org="" p="" 85900=""/>:
I've done MeDIP-seq (paired-end, 50 bp reads) with 18 samples from three groups and I would like to compare them using MeDIPS.
For mapping, I used bwa.
From other questions in here, MEDIPS: Number of windows tested in Paired-end data set<https: support.bioconductor.org="" p="" 81098=""/>, I can see that there might be some issues with the argument 'bwa' in e.g. MEDIPS.createSet, and when I set bwa = T, I also get the following warning: "...processing of bwa alignment files for paired-end sequencing data is still under development".
I’m currently using R version: 3.3.1 and MeDIPS: 1.22.0
When I google, I can see that documentation for a newer version of MeDIPS, 1.23.2, have been made and that this version skip the parameter ‘bwa’. However, I’m not able to upgrade to this version through biocLite. Can you tell me, when will this become possible?
Until then, which settings can be recommend? Should I re-map using Bowtie?
Post tags: R, medip-seq, medips, bwa
You may reply via email or visit MeDIPS and use of the 'bwa' parameter