<html>
  <head>
    <meta content="text/html; charset=utf-8" http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <div class="moz-cite-prefix">On 11/26/2014 01:59 AM, David Mobley
      wrote:<br>
    </div>
    <blockquote
cite="mid:CAPCBi1_JsHg9+iybG2e9BnBQwk--FuCy-6QSk2PF7_jms3Jmkg@mail.gmail.com"
      type="cite">
      <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
      <div dir="ltr">Hi,
        <div><br>
        </div>
        <div>Sorry for the delay. Our computer cluster was down for
          maintenance and I just got access to the files, at least for
          now.</div>
        <div><br>
        </div>
        <div>We do not specify sc-coul in our mdp files, which
          apparently means it is set to its default value (sc-coul =
          no). </div>
        <div><br>
        </div>
        <div>Our normal procedure is to first turn off charge
          interactions, then turn off vdW interactions separately
          after-the-fact, and we have never needed sc-coul with this
          procedure, so we have never used it. (Our protocol dates to
          before this was added). </div>
        <div><br>
        </div>
        <div>I'm guessing that this particular case, where there is a
          perturbed atom with a charge but zero LJ parameters, is a
          special case where there is "unexpected" behavior unless
          sc-coul is used?</div>
      </div>
    </blockquote>
    No, this should always work.<br>
    Lets continue this discussion on redmine.<br>
    <br>
    Cheers,<br>
    <br>
    Berk<br>
    <blockquote
cite="mid:CAPCBi1_JsHg9+iybG2e9BnBQwk--FuCy-6QSk2PF7_jms3Jmkg@mail.gmail.com"
      type="cite">
      <div dir="ltr">
        <div><br>
        </div>
        <div>I'll upload files shortly to Redmine.</div>
        <div><br>
        </div>
        <div>Thanks.</div>
      </div>
      <div class="gmail_extra"><br>
        <div class="gmail_quote">On Thu, Nov 20, 2014 at 3:49 AM, Berk
          Hess <span dir="ltr">&lt;<a moz-do-not-send="true"
              href="mailto:hess@kth.se" target="_blank">hess@kth.se</a>&gt;</span>
          wrote:<br>
          <blockquote class="gmail_quote" style="margin:0 0 0
            .8ex;border-left:1px #ccc solid;padding-left:1ex">
            <div text="#000000" bgcolor="#FFFFFF">
              <div>Hi,<br>
                <br>
                But this can not explain your crashes. If all
                interactions with a perturbed atom without LJ are
                soft-cored, there is never a singularity, since all
                interactions are soft-cored.<br>
                <br>
                Just to be sure, what have you set sc-coul to? This
                parameters is causing lots of headaches and I still have
                not fully understood the behavior. We need clear
                documentation here. I will add this soft-coring
                selection to the manual.<br>
                <br>
                Cheers,<br>
                <br>
                Berk
                <div>
                  <div class="h5"><br>
                    <br>
                    On 11/20/2014 12:37 PM, <a moz-do-not-send="true"
                      href="mailto:dmobley@gmail.com" target="_blank">dmobley@gmail.com</a>
                    wrote:<br>
                  </div>
                </div>
              </div>
              <div>
                <div class="h5">
                  <blockquote type="cite"> <span>
                      <div>Hi, all,</div>
                      <div><br>
                      </div>
                      <div>OK, to follow up on this - I think this is
                        incorrect at least in part, and explains a crash
                        I’ve had on my list of things to track down.
                        Specifically, if I simulate multiple carboxylic
                        acids in solution and attempt a hydration free
                        energy calculation (as a specific example, let’s
                        take a case I’ve done where I have neutral
                        benzoic acid at a finite concentration in water,
                        and I am attempting to compute the free energy
                        of turning one of the benzoic acid solutes into
                        dummy atoms) I simply CANNOT get the calculation
                        to run stably, even with reduced time steps,
                        etc. In every respect it behaves like the type
                        of crash I would expect if I had charged sites
                        overlapping one another, but I had been unable
                        to figure out how this could possibly happen. I
                        WAS, however, able to determine that if I added
                        nonzero LJ parameters to the hydroxyl hydrogen
                        on the carboxylic acid (which is a GAFF hydrogen
                        and hence had zero LJ parameters) the crashes go
                        away. </div>
                      <div><br>
                      </div>
                      <div>I believe this is explained by your statement
                        about “for atoms without LJ, all perturbed
                        interactions are soft-cored”. Specifically, I am
                        turning off the solute (carboxylic acid) and so
                        its perturbed interactions with the REMAINING,
                        non-perturbed solutes are soft-cored, meaning
                        that charge sites can overlap. At least, that’s
                        my reading of what you’re saying and it would
                        perfectly explain the crashes I’m seeing.</div>
                      <div><br>
                      </div>
                      <div>I’ve also seen essentially identical crashes
                        I was unable to understand for binding free
                        energy calculations in a carboxylic-acid-rich
                        host-guest system, presumably again because the
                        REMAINING hydroxyl hydrogens on the host end up
                        being able to overlap with charge sites on the
                        guest which is being turned into dummies. Again
                        I can “fix” it by adding nonzero LJ parameters
                        to the hydrogens on the host, though this is
                        incorrect since it is not “allowed” by the force
                        field.</div>
                      <div><br>
                      </div>
                      <div>Unless I’m missing something, this is a
                        substantial mistake, and clearly a better
                        solution is needed. </div>
                      <div><br>
                      </div>
                      <div>Thanks so much,</div>
                      <div>David</div>
                      <div><br>
                      </div>
                    </span>
                    <div> <br>
                      <span
style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13px;line-height:normal;background-color:rgb(255,255,255)">David
                        Mobley</span><br
style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13px;line-height:normal">
                      <a moz-do-not-send="true"
                        href="mailto:dmobley@gmail.com"
                        style="font-family:arial,sans-serif;font-size:13px;line-height:normal"
                        target="_blank">dmobley@gmail.com</a><br
style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13px;line-height:normal">
                      <span
style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13px;line-height:normal;background-color:rgb(255,255,255)"><a
                          moz-do-not-send="true" href="tel:949-385-2436"
                          value="+19493852436" target="_blank">949-385-2436</a></span><br
style="color:rgb(34,34,34);font-family:arial,sans-serif;font-size:13px;line-height:normal">
                    </div>
                    <br>
                    <br>
                    <div class="gmail_quote">
                      <p>On Wed, Nov 19, 2014 at 3:36 AM, Berk Hess <span
                          dir="ltr">&lt;<a moz-do-not-send="true"
                            href="mailto:hess@kth.se" target="_blank">hess@kth.se</a>&gt;</span>
                        wrote:<br>
                      </p>
                      <blockquote class="gmail_quote" style="margin:0 0
                        0 .8ex;border-left:1px #ccc
                        solid;padding-left:1ex">
                        <div>Hi,<br>
                          <br>
                          Soft-core is not decided on an atom basis. A
                          pair interaction is soft-cored when at least
                          one of the two atoms is perturbed and at least
                          one of the atoms has c12=0 at lambda=0 or
                          lambda=1. So for atoms with LJ, interactions
                          are only soft-cored when strictly necessary.
                          For atoms without LJ, e.g. many hydrogens in
                          many force fields, all perturbed interactions
                          are soft-cored. For such atoms soft-coring is
                          necessary when connected to a heavy atom that
                          is decoupled. So the only cases where some
                          interactions are soft-cored that are not
                          strictly necessary is in a system where some
                          atoms are decoupled and others are modified,
                          then interactions with hydrogens connected to
                          modified atoms would not need to be
                          soft-cored.<br>
                          <br>
                          Cheers,<br>
                          <br>
                          Berk<br>
                          <br>
                          On 11/18/2014 11:45 PM, David Mobley wrote:<br>
                        </div>
                        <blockquote type="cite">
                          <p dir="ltr">Hi,</p>
                          <p dir="ltr">I wanted to ask concerning
                            soft-core potentials in GROMACS: How is it
                            determined which atoms are soft core?
                            Specifically, some other code bases allow
                            the option to select specific atoms which
                            are going to be treated with LJ soft core
                            potentials, while other atoms are not
                            treated in this way and maintain their
                            normal potential or a linear mix of their A
                            and B state potentials. In GROMACS, there is
                            no option to select this that I am aware of,
                            so I'm assuming that soft core is applied to
                            all atoms, or to all atoms which have
                            different A and B states or different A and
                            B state LJ parameters?</p>
                          <p dir="ltr">Thanks!<br>
                            David<br>
                          </p>
                          <br>
                          <fieldset></fieldset>
                          <br>
                        </blockquote>
                        <br>
                      </blockquote>
                    </div>
                    <br>
                  </blockquote>
                  <br>
                </div>
              </div>
            </div>
          </blockquote>
        </div>
        <br>
        <br clear="all">
        <div><br>
        </div>
        -- <br>
        <div class="gmail_signature">David Mobley<br>
          <a moz-do-not-send="true" href="mailto:dmobley@gmail.com"
            target="_blank">dmobley@gmail.com</a><br>
          949-385-2436<br>
        </div>
      </div>
    </blockquote>
    <br>
  </body>
</html>