Correction for Tris buffers

[noadv...@googlemail.com]

Hi,

first let me thank you for creating such a nice tool for computing melting temperatures. 

We are using MELTING in our multiplex primer design tool openPrimeR and noticed a problem with the melting temperature computations with MELTING 5.1 if a Tris buffer is specified. The computed Tms seem to be way too low if they are corrected for Tris. Assume that "sequencefile.txt" contains a single oligo and its complement, namely

ACTGAACGGGATTAAACTGGAA TGACTTGCCCTAATTTGACCTT 

With typical ion concentrations but without Tris correction, the result computed via

melting-batch -E Mg=0.0015 -E K=0.05 -H dnadna -P 0.0000001 ~/sequencefile.txt

is 51.4 degrees Celsius. But, if I specify a Tris buffer via

melting-batch -E Mg=0.0015 -E K=0.05 -E Tris=0.005 -H dnadna -P 0.0000001   ~/sequencefile.txt 

is 20.97 degrees Celsius. I tried to obtain higher Tms by experimenting with the available ion corrections and found that "-ion owc2004" leads to the best result but the estimated temperature seems to be too low still. Do you have any idea why that could be? Currently, we are considering simply disregarding the Tris concentration.

I also wanted to make you aware that the documentation at https://www.ebi.ac.uk/biomodels-static/tools/melting/melting5-doc/melting.html is misleading. The documentation suggests that [Tris+] rather than [Tris] should be provided via the CLI. Indeed, however, the tool expects the Tris buffer concentration. So the following line in the documentation

"Enters the different ion (Na, Mg, Tris, K) ..."

should not include Tris.

Kindly looking forward for your feedback

 Matthias Döring

[Nicolas Le Novere]

Dear Matthias,

Sorry for the long delay of response. I am no longer in academia. Melting5 has been developed by a range of people, including Marine Dumousseau-Sivade, Nicolas Rodriguez and Piero Dalle-Pezze. I am not certain any of them is able to work on it.

But the good news is that MELTING5 is Free Software, released under the GPLv3. You are most welcome to debug and modify the code. At the moment, I do not believe there is a Github repo (I might be wrong. Nico, Piero, if you read this, could you pitch in?). But you can find the latest public sources at:

https://sourceforge.net/projects/melting/files/meltingJava/melting5/

Thank you for noticing the error in the documentation. I will see if I can modify it.

[Nicolas Le Novere]

Dear Matthias,

Are you certain for the Tris? In Melting 4, one entered the Tris buffer concentration using the -t argument. However, in Melting 5 one must enter the concentration of Tris+, using the -E argument (which should be 0.5 the concentration of buffer).

At least, this is what the documentation states.

https://www.ebi.ac.uk/biomodels-static/tools/melting/melting5-doc/melting.pdf

pages 5, 6, 59

On Monday, November 5, 2018 at 2:24:33 PM UTC, noadv...@googlemail.com wrote:

[Nicolas Le Novere]

In the addition of packages, all the source code is available at:

https://sourceforge.net/p/melting/code/HEAD/tree/