How can I compute DNA methylation levels around splice junctions?

Question

[thefirstrealace]

Hello everybody,

I am still new to the field of computational epigenetics, so I need some help with the following task(s):

I study applied bioinformatics and in the context of my master thesis, I need to compute methylation levels around splice junctions. I need to output it in a format that I have never seen before. I did some research about the format, but I couldn't find anything about it. 'The format seems to be similar to fasta, but instead of a sequence (after the header starting with >), it provides methylation levels in a tab-seperated manner, and I honestly don't know what DSQ stands for. A small part of a methylation track is given below is given below:

>chr1:142346773:142346881:+@chr1:142380702:142380810:+@chr1:142404277:142404426:+_expu=400_expd=200_bsz=20_part=0
DSQ    18.5594    18.5594    18.5594    8.22605    18.5594    31.9349    36.4521    36.4521    33.8659    18.5594    8.22605    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0    0
>chr1:58852214:58852582:+@chr1:58878691:58878806:+@chr1:58880759:58881091:+_expu=400_expd=200_bsz=20_part=0
DSQ    0    0    0    0    0    0    0    0    4.50575    ...

This format is recognized by a newly developed flexible self-organizing map for DNA methylation analysis (or other digitized epigenetic signals). The paper describing the software is freely accesible here. Unfortunately besides a paper describing this software, the authors provide a 3-page-quick-start-manual, which doesn't tell much about this format shown above, but maybe someone here has seen this format before and can explain me the anatomy of it.

What I have done so far:

1. I downloaded RNA-Seq runs from human spleen sample provided by NIH Roadmap Epigenomics Project. The GEO accession is GSM1010976.
2. I used TopHat splice junction mapper in order to determine splice junctions and therfor used hg19 as reference genome.

I need to compute:

• The methylation levels in the range -200nt/+200nt to the left/right of these splice junctions respectively
• I need them in 20nt intervals. These DSQ values seen in the above example represent the (normalized?) methylation levels within a 20nt bin

I also found the data of whole genome BS-Seq experiment which was done for the same spleen sample. The GEO accession is GSM983652. I considered the following possibilities:

1. If I understand correctly, the provided wig-file already contains methylation data. If that is the case, I would like to use the already existing methylation data. Is there a tool to extract methyation data out of a wig file? As I said before I need the cytosine methylation levels near splice junctions and I need them to be exported in the format shown above.
2. If option 1 doesn't work, which tool should I use to analyse the provided BS-Seq data? And again: How can I export them in the format shown above?

I hope that somebody can help me with these tasks.

Best regards

thefirstrealace