This page discusses the technical procedures required to run a folding molecular dynamics simulation (using the AMBER family of force fields) in methanol or TFE. In both cases, we examine the case of homogeneous solvent as well as a mixture with water). If you wanted to do a proper job (and not this hack) you can take ideas from here.

To use methanol as solvent is a piece of cake. All you have to do is change one word in the 'solvateBox' command of your leap script :

source leaprc.ff99SBHBildn
loadamberparams frcmod.ionsjc_tip3p
ftzpep = loadPdb aligned.pdb
check ftzpep
solvateBox ftzpep MEOHBOX 11.0 iso            <=========== 'MEOHBOX' is the only difference
addions ftzpep Cl- 2
saveAmberParm ftzpep ftzpep.prmtop ftzpep.crd
savepdb ftzpep ftzpep.pdb

If you want to prepare a methanol/water mixture, read further below for a hack that you may find useful.

---

With TFE things are more complex (and I have taken a rather cavalier approach on how to equilibrate the new solvent). First thing first. The parameters for TFE were taken from here and were converted to an AMBER OFF file using the scripts provided in the same page. The resulting file TFE.off has been placed in the LEAP subdirectory of AMBERTOOLS.

Now comes the cavalier bit :

Normally we would have to properly prepare a pre-equilibrated box of the new solvent. But since what will follow is a folding simulation (that starts from the unfolded state), you can get away by starting with a lattice placement of solvent and then let heating take care of equilibration. For this to work you need to significantly increase the extent of heating phase. For a pure TFE simulation use this LEAP script

source leaprc.ff99SBHBildn
loadoff TFE.off
ftzpep = loadPdb aligned.pdb
check ftzpep
solvateBox ftzpep TFE 15.0 iso
addions ftzpep Cl- 2
saveAmberParm ftzpep ftzpep.prmtop ftzpep.crd
savepdb ftzpep ftzpep.pdb

which gave an initial system that looked disgusting :

but after extensive heating at 360K got much better :

---

If you think that this was bad, wait to see how you can prepare a water/TFE mixture. First you need to know the ratio of water/TFE molecules you need in order to get a given by volume percentage. Get the numbers from this paper (table III). Then, you systematically modify the following LEAP script until you get the correct composition (with respect to the number of TFE/water molecules) :

source leaprc.ff99SBHBildn
loadamberparams frcmod.ionsjc_tip3p
loadoff TFE.off
ftzpep = loadPdb aligned.pdb
check ftzpep
solvateBox ftzpep TIP3PBOX 5.5 iso                    <============
solvateBox ftzpep TFE 7 iso                           <============
addions ftzpep Cl- 2
saveAmberParm ftzpep ftzpep.prmtop ftzpep.crd
savepdb ftzpep ftzpep.pdb

The resulting system is again disgusting (you can see the inner box of equilibrated water surrounded by the lattice of TFE) :

Extensive heating at 360K significantly improved the situation :

---

It is too early to say, but first tests with a peptide known to form a stable α-helix in the presence of 50% TFE look quite promising (noting that this is a 4 μs adaptive tempering simulation with a temperature range of 280-380K) :

---

Using DMSO goes more-or-less along the same disgusting lines as with TFE : you prepare a system with the peptide in a horrible square lattice system, and then let heating take care of equilibration.

The leap file should look like this :

source leaprc.ff99SBHBildn
loadamberparams frcmod.ionsjc_tip3p
loadamberparams dmso.frcmod
loadamberprep dmso.prepi
AAQAA = loadPdb aligned.pdb
check AAQAA
solvateBox AAQAA DMS 10.0 iso
saveAmberParm AAQAA AAQAA.prmtop AAQAA.crd
savepdb AAQAA AAQAA.pdb

The files dmso.frcmod and dmso.prepi come directly from

https://github.com/choderalab/ambermini/tree/master/share/amber/dat/leap/pol_solvent_database.

The images before and after are :

Using the same trick described above for TFE, you can also build mixed DMSO/water systems :

---

Results from 1 μs long simulations of the (AAQAA)3 peptide :