17 replies to this topic

[gungrave]

Hello everyone I'm new here and I was wondering if I could get some assistance.
    In my research of the literature I thought the differences between the two methods were based on the proximity of the primers to the CpG sites and that BSP primers were designed without including CpG dinucleotides.
   I was wondering if someone could clear it up for me and what are the pros and cons of each method. Thank you very much in advance

This post has been promoted to an article

[methylnick]

Hi Gungrave,

I assume by BSP you mean bisulfite PCR and sequencing. If so, the advantages of this over MSP are:

A greater number of CpG residues analysed for methylation when compared with MSP (MSP is only one and that is identified with a methylation specific primer). Bisulfite sequencing looks at every CpG residue across the whole amplicon.

I would also say that primer optimisation is crucial for an MSP assay to work, it is not so crucial for BSP because you will be sequencing through the amplicon anyway.

MSP has it advantages in that it is far more quicker than BSP in terms of hands-on lab time.


hope this helps. I would be interested to see what other people think about this also.

Cheers

nick

[gungrave]

Thanks for the reply. I always thought that with MSP one could include a number of Cpg dinucleotides in the region being amplified especially in the 3' end primer design to make the primer more specific.
   Does this mean we only look at one CpG dinucleotide with MSP and if we want to observe several then use BSP?
   My confusion started today when I ran across a paper that described the MSP they conducted by using restriction enzyme isoschizomers and amplifying the region that the enzymes cut. Thanks once again.

[methylnick]

gungrave, on Feb 28 2005, 11:29 PM, said:

   Does this mean we only look at one CpG dinucleotide with MSP and if we want to observe several then use BSP?
   My confusion started today when I ran across a paper that described the MSP they conducted by using restriction enzyme isoschizomers and amplifying the region that the enzymes cut. Thanks once again.

I suppose it is possible to look at multiple CpGs with MSP primers. Just clone and sequence the amplicons.

MSP has advantages in that you are able to gauge the methylation status by detecting the amplicon. An MSP primer usually ends to discriminate either a methylated or unmethylated C so it is a really quick way to assaying for methylation.

I think you are referring to COBRA which stands for combined bisulfite and restriction analysis. This is like the midway between MSP and BSP, COBRA is limited to the recognition site of the restriction enzyme, so you can potentially assy for more than one CG residue, but you are not assaying all of them.

Cheers

Nick

[labtechie]

I agree with Nick-- MSP, even combined with a BstI restriction of the amplicon, is called site specific; it only lets you look at however many CpGs are in your primers and in a BstI site if you are lucky enough to have one.  I wanted to use this method as well to avoid sequencing and cloning but in the end I have determined that sequencing is the only thing I trust and I tend to scoff at papers who use MSP as it is so prone to artifact.  Check out my agarose bead protocol in the "bisulfite treatment" thread, pinned above.

Good luck,
Ellen

[gungrave]

Thank you all for for your replies, there's one more thing and I hope I'm not beating a dead horse, bear with me if so   .

The differences I see in MSP and BSP come from the primer design where MSP primers are designed with CpG(better in the 3' end) to make them more specific.

BSP on the other hand, primers are designed to encompass the CpG without including them in the design of the primers.

Does that sound about right? Thank you all very much

[labtechie]

That is right-- for BSP,  you want NO CpGs in your primers, and convert all non-CpG Cs into Ts in the forward primers (you figure out the reverse).   BSP should amplify all converted strands regardless of methylation status, then it's the sequencing that matters.  But do you see how you get a much better picture of the whole island this way?

[gungrave]

gotcha, thanks for clearing that up

[pcrman]

I second Ellen about MSP. If PCR conditions and primer design are not optimized, its result is prone to artifact.  I simply won't believe any PCR method that relies on one nucleotide to avoid mispairing, even the nucleotide is at the 3' end of the primer.

[Ming]

pcrman, on Mar 1 2005, 05:03 PM, said:

I second Ellen about MSP. If PCR conditions and primer design are not optimized, its result is prone to artifact.  I simply won't believe any PCR method that relies on one nucleotide to avoid mispairing, even the nucleotide is at the 3' end of the primer.

I very agree about that...

[methylnick]

gungrave, on Mar 1 2005, 08:26 AM, said:

BSP on the other hand, primers are designed to encompass the CpG without including them in the design of the primers.

Hi everyone,

I might add that for BSP, if it can not be avoided then primers may have CpGs in them, ideally you want them at the 5' end, and to ensure non-bias, make the CG a YG, ie have a degenerate base at the site of potential methylation or non-methylation.

Cheers and good luck!!

nick

[pcrman]

Quote

I might add that for BSP, if it can not be avoided then primers may have CpGs in them, ideally you want them at the 5' end, and to ensure non-bias, make the CG a YG, ie have a degenerate base at the site of potential methylation or non-methylation.

There is a bisulfite primer design tool called bisearch which can design such degenerate primers at bisearch.enzim.hu

[et2b]

pcrman, on Mar 2 2005, 07:11 AM, said:

Quote

I might add that for BSP, if it can not be avoided then primers may have CpGs in them, ideally you want them at the 5' end, and to ensure non-bias, make the CG a YG, ie have a degenerate base at the site of potential methylation or non-methylation.

There is a bisulfite primer design tool called bisearch which can design such degenerate primers at bisearch.enzim.hu

Hi, I know most people from cancer field etc. etc. love MSP, but MSP is quite bullshitty. Just one example: Check the papers from Lillycrop et al. in the british journal of nutrition. Lay their rat papers beside eachother. First thing to notice: what the F%%$$#....how can you publish the same results 3 fucking times in the same %^$#@ journal.
Than sequencing 1.6 % difference between 2 groups: MSP says >80%.....Just love MSP dont you?  

And what the &$#%^& why measure one CpG if you can run a ABI trace and measure several CpGs?????? semi-quantitatively(with the right software).
come one people: Science is a emperical thingy....so measure the bloody things and dont use proxy things

[battlebee]

et2b, on Mar 2 2009, 03:26 PM, said:

And what the &$#%^& why measure one CpG if you can run a ABI trace and measure several CpGs?????? semi-quantitatively(with the right software).

Could you please recommend a software tool for a semi-quantitative analysis

Thanks a lot

[et2b]

battlebee, on Mar 23 2009, 03:13 PM, said:

et2b, on Mar 2 2009, 03:26 PM, said:

And what the &$#%^& why measure one CpG if you can run a ABI trace and measure several CpGs?????? semi-quantitatively(with the right software).

Could you please recommend a software tool for a semi-quantitative analysis

Thanks a lot

Hi, sorry I am so late:

ESME

best wishes