DMR/CpG

Question

[emiliomastriani]

Dear all, I got a couple of fastq.gz files from my lab. These files have been sequenced using the Illumina Novaseq 6000 and contain data from methylated DNA captured by MBD enrichment method. I already accomplished the preliminary analysis (trimming/statistics, create index file to GRCh38, duplicate tagging, generation of vcf/cov file) using two different approaches: one based on BISCUIT/SAMTOOLS and another one based on nf-core/methylseq (BISMARK). I obtained the vcf file (together with the bed and bam) from the first strategy, and I the cov file (bismark format, together with the bam etc) from the second one. Now, considering I am a newbie in methylation analysis, I have to:

1. identify the DMR and CpG;
2. make prediction of the target gene methylation;
3. identify the region/promoter in those genes;
4. try to use some DBs (TCGA??) for mapping.

I already tried to use GeneDMR and dmrseq, but I got lost ... Please, can someone help me to understand how to go forward? Which tool to use? Maybe there is some step-by-step guide suitable to my case... Any advise will be very welcome. Thank you

[emiliomastriani]

Is there anybody can help me?

[emiliomastriani]

I moved ahead with my analysis. Coming my data from Illumina Novaseq 6000 and being the methylation sequencing obtained with the capture based method using Methyl Binding Domain (Human MBD2A), I opted to use the nf-core/methyl-seq (Methyl-Dackel) pipeline. It worked perfectly and I got the results from MethylDackel. Now I have to check for a tool (R, python or similar) to give me back the DMRs and DMCs, suitable to my data (I repeat: the methylated regions have been gain with captured based method using Methyl Binding Domain). Can you suggest me someone? Thank you very much.

Answer 1

[Phil Ewels]

I'd recommend having a read over the course material from the Babraham Institute about bisulfite analysis (written by the guys behind Bismark): https://www.bioinformatics.babraham.ac.uk/training.html#bsseq

[emiliomastriani]

Hello Phil, thank you for your answer. Anyway, I moved ahead with my analysis. Coming my data from Illumina Novaseq 6000 and being the methylation sequencing obtained with the capture based method using Methyl Binding Domain (Human MBD2A), I opted to use the nf-core/methyl-seq (Methyl-Dackel) pipeline. It worked perfectly and I got the results from MethylDackel. Now I have to check for a tool (R, python or similar) to give me back the DMRs and DMCs, suitable to my data (I repeat: the methylated regions have been gain with captured based method using Methyl Binding Domain). Can you suggest me someone? Thank you very much.