Hi Francesco,

I think the problem is not reliability but the SD values associated with most methods that measure total 5-methyl C because when you extrapolate their values, they exceed the whole number of CpG islands of all single gene promoters…. Most genomic methylation is concentrated on repeat DNA, etc….Unless there is a substantial changes in the order of several millions methylation sites, I am not sure you can detect anything outside the standards deviation values.

You should also consider that you may have false negative results if there is gain of methylation from some genomic region and loss of methylation from others which may suggest to you that there is no methylation change upon cell stimulation.

Thanks for the answer Abdelhalim.

Especially the compensating effect of gain & loss of 5mC was something I had already taken into account.

Could you suggest any technique other than NGS to address this type of investigation?

Thanks again for your help.

Hi Francesco,

I think you should perform your experiment as you have planed. I only wanted you to have in mind some parameters. It is always interesting if you can find some specific stimulation which is capable to induce global genomic DNA methylation changes. Your next step will be to locate their genomic loci...

You can use the Infinium HumanMethylation450 BeadChip Kit as an alternative to NGS. In the past we have used the 27K beads array with negative results. We could find very important data from NGS (still to be published). I think if you can gather some interesting preliminary data you can always find a collaborating lab which has access to the NGS facility and most importantly to the bioinformatics.

The last resort is a candidate approach. The literature is full of them

Hi Francesco,

I am also interested in this kit. Have you run your trials? How successful were you?

Regards,

Sophia

Hi folks,

I have been attempting to use the Imprint DNA methylation Quantification Kit (Sigma Aldrich) which is very similar in practice to the Epigentek kit you were discussing. I have been finding that the results I get from this kit are unusable. There is extremely high variation among technical replicates and I obtain very high background readings (in excess of my sample readings) in spite of following the instructions to the letter and further instruction from the Sigma technical dept to drop the colour development time to a minute or two (no more).

How have you all gotten on with your experiments and this type of kit? Would be very grateful for any tips :)

Many thanks

Nicki

Have just run a colorimetric Epigentek plate. Pretty high variability between replicates but the standard curve was ok. Our absolute  %CpG methylation results ranged from 0.8-2.6% -- for mammalian brain we were expecting much much higher (60-80% as you would get with a LUMA-type analysis).

These are similar to what Epigentek lists on their product website for a tissue type comparison but perhaps I have misunderstood what exactly they are are measuring. Is it only a subset of CpGs? Any insights would be appreciated!

Thanks

Dear all,

thanks for your feedback. I must admit I dropped this method after some unsuccessful tries. We managed to achieve quite some precision on technical replicates, although biological ones (inter-experimental variability) were rather different.

Moreover, as reported in the first comment, only substantial changes can be actually assesed.

In the end, we opted in favour of NGS methods, which have become rather affordable.

Hi all,

I too have tried out the ELISA method using bird DNA. We found quite a good standard curve, however there was huge variation between our sample duplicates. All of our samples were said to show less than 1% methylation. I wonder if any one knows of any publications / sources that say anything about the unreliability of this kit!

Thanks

I'm so glad to hear that others have experienced unreliable results from the Epigentek DNA Methylation kits. I've run 4 plates and the variability between sample duplicates and between plates is unbelievable. I'm in contact with the company now to see if there's some trick that I'm missing, but thus far, I haven't been impressed with the kit's performance. 

I would definitely recommend the 5mC kits from either Zymo or Enzo. They both use the same principle and work great but make sure the kit is handled and delivered with care with the antibodies in dry ice as they are extremely sensitive. I have also used kits from Epigentek and abcam and I have not been able to get good replicate data from them.  

Hi all. We would like to inform everybody that we have recently launched our new MethylFlash "Easy" kit, a fourth generation product of our global DNA methylation ELISA-based method, currently available with a 25% product launch discount to US customers (contact a distributor for a special discount if outside of the US):

https://www.epigentek.com/catalog/methylflash-global-dna-methylation-mc-elisa-easy-kit-colorimetric-p-5370.html

With the previous MethylFlash kit, we were aware of concerns of occasionally high intra-assay CV% some users experience. This was particularly tricky for certain samples given the low amounts of the target of the assay (specifically, methylated DNA). Thus we worked hard to try to further refine and improve replicate data. This new MethylFlash Easy kit introduces a few tweaks to the reagents, as well as a change in the plate layout, that will reduce CV. 

Also, Danielle Dillon, please feel free to reach out to me or an account rep and I would be happy to send you a free 48-reaction size kit of the new MethylFlash Easy kit. This is a finalized, public version -- not a beta test version -- so we are confident that it will perform well for you. Good luck!

Hello,

Here you can find a comparison of four different ELISA-based DNA methylation kits:

https://www.karger.com/Article/Abstract/486201