MeDIP-seq - input sample needed?

[ETHANol]

I've prepared some MeDIP-seq libraries and they appear pretty good and ready to be run on the HiSeq. My question is, do I need to run input control libraries like we do with ChIP-seq or is the fragmentation of the genome and subsequent amplification so even that it is not necessary?

I have samples from several cell lines and was thinking of just running one input control?

[msheldon]

Input controls are not a bad idea, what did you use to fragment?
Are you using methylated or non-methylated adapters?

[ETHANol]

I used the Covaris S2 and their suggested settings using microTubes.

For adapters, I used adapters with the TruSeq sequences synthesized by IDT and non-methylated. I am aware that the TruSeq adapters from Illumina are methylated.

My libraries have about a 50-fold difference in signal between what should be a methylated region and a CpG-less sequence using qPCR and normalized to input. From what I understand this is good enrichment for MeDIP. Do others see better enrichment?

[msheldon]

that will work. we sequence material with 40-60 fold greater signal. let me know how it turns out.

[helitron]

My libraries have about a 50-fold difference in signal between what should be a methylated region and a CpG-less sequence using qPCR and normalized to input. From what I understand this is good enrichment for MeDIP. Do others see better enrichment?

hey,
I always compare methylated regions to unmethylated regions of similar CpG density to measure enrichments - that way you avoid getting into sonication artifacts. Especially if you are testing your libraries and not the IP material for enrichment.
E.g. use a methylated CpG island vs an unmethylated one (take for example a housekeeping gene promoter). Don't know what cell type you're using but methylated CG islands are mostly at sperm specific promoters or imprinting control regions. Even if only one allele is methylated, it will still give you a good signal to measure enrichment. Stay away from repeats and transposons.