[meltsim] methylation issue

[Kevin Hollevoet]

Hi proffesor Marx!

i looked it up in the archieve, and found a question that concernded a  
difference in meth of a C, so the CH3 group itself. but that is not  
what we are using here. i'm sorry i didn't more specify the research  
method.

i was planning on doing it with bisulphite modified sequences. in that  
way meltsim would recognize the C>T differences, just like it would  
give a different curve for any mutation or difference. so if we  
analyze a methy seq, every C in a CpG stays a C, but onmethylated CpG  
transform into TpG,

that way you can analyse the methylation status on the hand of the  
melting curves. and find significant clustering in the difference  
plots, some research has been done on this issue, (tomasz K. wojdacz  
2007) but here they worked with fully methylated or fully onmeth  
sequences, and mixed them to become a certain % of methylation.

but i 'd like to check if we can use HRM for a methylation status  
check, even if the patron of methylation is somewhat heterogeneous in  
cellines (what is sometimes the case is some of our NB samples). we  
think this is not the case,and that it is impossible to become some  
kind of significant clustering. so in some way say something else than  
tomasz K. wojdacz  et all. and in order to give that idea some support  
i' d lik to check this with some kind of simulation,

meltsim gave me already some impression with the melthpieks, but  
normalized difference plots simulation would be better (that's the  
reason of my last email of today)

i don't think i need extra parameters in meltsim for this kind of simulation?

but please, feel free to correct me if i'm wrong!

kind regards,
kevin







Citeren "Marx, Kenneth" <Kenneth_Marx at uml.edu>:

> Hi Kevin,
>
> I don't know how you plan to do the study, but MELTSIM is not
> parameterized for melting methylated bases. This is something that Prof.
> Richard Blake, who developed and parameterized MELTSIM did not do. I
> asked him years ago about this issue of melting methylated bases.
>
> Good luck, Prof. Ken Marx
>
> -----Original Message-----
> From: meltsim-bounces+kenneth_marx=uml.edu at bioinformatics.org
> [mailto:meltsim-bounces+kenneth_marx=uml.edu at bioinformatics.org] On
> Behalf Of Kevin Hollevoet
> Sent: Wednesday, August 15, 2007 12:10 PM
> To: meltsim at bioinformatics.org
> Subject: [meltsim] problem running MELSTIM
>
> hello,
>
> i'm kevin hollevoet, thesisstudent at the university of Ghent
> (belgium) at the medical research building. we are planning to use
> meltsim to preform some large scale simulations to test the
> sensitivity of HRM towards single CpG's methylation differences in NB
> cell lines.
>
> but the problems we are facing are the same like Marko Pavlekovic
> wrote to you some time ago (for the record: i'm using window XP to run
> MELTSIM)
>
> so i tried, just like marko, to change all '.' into ','. even in the
> output files,  when i only selected the meltcurse, the graph was not
> what i would expect. other options they gave also errors for the use
> of E09 (meaning 10^9) in the output file. and that fault (i dont even
> know what the fault is) is obviously not that easy to change in an
> output file.
>
> so is there already some other version on the market or the web?
> perhaps this is some european problem with the configuration of the
> windows XP here? as there are a lot of articles citing your program,
> even recently, it should work somehow perfectly well, no?
>
> greetings
>
> kevin
> _______________________________________________
> meltsim maillist  -  meltsim at bioinformatics.org
> https://bioinformatics.org/mailman/listinfo/meltsim
>