<div dir="ltr"><div><div><div><div><div><div><div>Hi Berk,<br><br></div>Yes, I think I was calling that after the coordinate communication. The place where I code it in is after the command:<br><br>   if (bDoForces)<br>   {<br>      post_process_forces(fplog, cr, step, nrnb, wcycle,<br>        top, box, x, f, vir_force, mdatoms, graph, fr, vsite,<br>        flags);<br>   }<br><br></div>After that I use a <br></div>if(bNS){<br></div>      if (DOMAINDECOMP(cr)){<br><br></div>to make sure it is done every neighboring list step.<br><br></div>Is this correct? BTW, I was coding that in gromacs4.6.3 version (this is for historical reason because I have coded other unrelated things in other files).<br><br></div>I have checked that the ga2la_get_home in the first loop find all ai atoms, but the second loop ga2la_get missed some aj atoms. One of the reasons I guess is because the missing atom pairs 
are too far apart (but less than 6 angstroms) in Cartesian space from 
each other, so that even by including all cells of the home node it still could not find it. That could be a problem, but I don&#39;t know exactly how -dd find this pair partner in its neighboring cells. <div><div><div><div><div><div><div><br></div><div>Thank you,<br></div><div>Xingcheng<br></div><div><div><div class="gmail_extra"><br><div class="gmail_quote">On Fri, Aug 28, 2015 at 11:28 AM,  <span dir="ltr">&lt;<a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se" target="_blank">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">Send gromacs.org_gmx-developers mailing list submissions to<br>
        <a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
        <a href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers" rel="noreferrer" target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers</a><br>
<br>
or, via email, send a message with subject or body &#39;help&#39; to<br>
        <a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a><br>
<br>
You can reach the person managing the list at<br>
        <a href="mailto:gromacs.org_gmx-developers-owner@maillist.sys.kth.se">gromacs.org_gmx-developers-owner@maillist.sys.kth.se</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than &quot;Re: Contents of gromacs.org_gmx-developers digest...&quot;<br>
<br>
<br>
Today&#39;s Topics:<br>
<br>
   1. Re: dhdl expanded ensemble<br>
      (Mirabzadeh, Christopher (<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>))<br>
   2. Re: Increase the scope of dd domains (Berk Hess)<br>
<br>
<br>
----------------------------------------------------------------------<br>
<br>
Message: 1<br>
Date: Fri, 28 Aug 2015 16:17:40 +0000<br>
From: &quot;Mirabzadeh, Christopher (<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>)&quot;<br>
        &lt;<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>&gt;<br>
To: &quot;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&quot; &lt;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;,<br>
        &quot;<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a>&quot; &lt;<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a>&gt;<br>
Subject: Re: [gmx-developers] dhdl expanded ensemble<br>
Message-ID: &lt;<a href="mailto:3686D140-78FF-4961-80B0-C13B05A36EF7@vandals.uidaho.edu">3686D140-78FF-4961-80B0-C13B05A36EF7@vandals.uidaho.edu</a>&gt;<br>
Content-Type: text/plain; charset=&quot;utf-8&quot;<br>
<br>
Thank you! I will look into redmine and I will also look at those variables closer.<br>
<br>
Thanks,<br>
<br>
-ChrisM<br>
<br>
Mirabzadeh, Christopher<br>
Graduate Research Assistant/Physics Instructor<br>
Department of Physics<br>
University of Idaho<br>
Moscow, Id<br>
<a href="tel:%28509%29339-5647" value="+15093395647">(509)339-5647</a><br>
<br>
<br>
<br>
&gt; On Aug 28, 2015, at 5:58 AM, Shirts, Michael R. (mrs5pt) &lt;<a href="mailto:mrs5pt@eservices.virginia.edu">mrs5pt@eservices.virginia.edu</a>&gt; wrote:<br>
&gt;<br>
&gt; Yeah, apologies, checking 5.1 in the in the debugger for a binding free<br>
&gt; energy change, then there are some differences between the dhdl output<br>
&gt; file when nstdhdl = 20, and the terms in the enerd-&gt;term dhdl components<br>
&gt; at step = 10 matched what the output was in the dhdl file when nstdhdl =<br>
&gt; 10.  So don&#39;t rely on that dhdl, since it is missing some terms.<br>
&gt;<br>
&gt; In any case, it would be relatively easy to change the existing machinery<br>
&gt; to force it do the full calculation dhdl calculation every step.  If you<br>
&gt; wanted to make it very easy, just set nstdhdl = 1, and introduce a new<br>
&gt; variable that controls the printing of dhdl to the file alone.<br>
&gt;<br>
&gt; ~~~~~~~~~~~~<br>
&gt; Michael Shirts<br>
&gt; Associate Professor<br>
&gt; Department of Chemical Engineering<br>
&gt; University of Virginia<br>
&gt; <a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a><br>
&gt; <a href="tel:%28434%29%20243-1821" value="+14342431821">(434) 243-1821</a><br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt; On 8/28/15, 3:11 AM, &quot;Berk Hess&quot; &lt;<a href="mailto:hess@kth.se">hess@kth.se</a>&gt; wrote:<br>
&gt;<br>
&gt;&gt; No. To be more precise dhdl is only (fully) calculated at each step<br>
&gt;&gt; where free energies are calculated, i.e. every nstdhdl steps.<br>
&gt;&gt; To find where the final values are stored, it&#39;s often easiest to look in<br>
&gt;&gt; mdebin.c, in this case at line 1244.<br>
&gt;&gt;<br>
&gt;&gt; Cheers,<br>
&gt;&gt;<br>
&gt;&gt; Berk<br>
&gt;&gt;<br>
&gt;&gt; On 2015-08-28 03:52, Shirts, Michael R. (mrs5pt) wrote:<br>
&gt;&gt;&gt; Dhdl is automatically calculated at every step if free energies are<br>
&gt;&gt;&gt; calculated. Look in sum_dhdl -- difference components are stored in<br>
&gt;&gt;&gt; enerd-&gt;term.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Unless it&#39;s a new development you would rather keep private for now, you<br>
&gt;&gt;&gt; should consider posting the proposed enhancement on redmine.  You would<br>
&gt;&gt;&gt; likely get some tips about how to do it in a way that could be more<br>
&gt;&gt;&gt; easily<br>
&gt;&gt;&gt; included in later gromcs versions.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Best,<br>
&gt;&gt;&gt; ~~~~~~~~~~~~<br>
&gt;&gt;&gt; Michael Shirts<br>
&gt;&gt;&gt; Associate Professor<br>
&gt;&gt;&gt; Department of Chemical Engineering<br>
&gt;&gt;&gt; University of Virginia<br>
&gt;&gt;&gt; <a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a><br>
&gt;&gt;&gt; <a href="tel:%28434%29%20243-1821" value="+14342431821">(434) 243-1821</a><br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; From:  &lt;Mirabzadeh&gt;, &quot;Christopher   (<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>)&quot;<br>
&gt;&gt;&gt; &lt;<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>&gt;<br>
&gt;&gt;&gt; Reply-To:  &quot;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&quot; &lt;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;<br>
&gt;&gt;&gt; Date:  Thursday, August 27, 2015 at 12:20 PM<br>
&gt;&gt;&gt; To:  &quot;<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&quot;<br>
&gt;&gt;&gt; &lt;<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&gt;<br>
&gt;&gt;&gt; Subject:  [gmx-developers]  dhdl expanded ensemble<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Hello,<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; What I&#39;m trying to do;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I&#39;m attempting to add functionality to expanded ensemble.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; What I&#39;ve done so far;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I&#39;ve added functionality to accept my mdp options and found pre-written<br>
&gt;&gt;&gt; functions to take advantage of.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Where I&#39;m stuck;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I need the derivative of the Hamiltonian with respect to lambda at every<br>
&gt;&gt;&gt; md step. I&#39;m having a hard time narrowing down this variable. I know<br>
&gt;&gt;&gt; it&#39;s<br>
&gt;&gt;&gt; calculated and I see in mdebin.c there is a line that says &quot;/* store_dh<br>
&gt;&gt;&gt; is<br>
&gt;&gt;&gt; dE */&quot;. Is the derivative &quot;store_dhdl[]=enerd-&gt;term[F_DVDL]&quot;?<br>
&gt;&gt;&gt;   Or, is this already being called somewhere in expanded.c that I can<br>
&gt;&gt;&gt; take<br>
&gt;&gt;&gt; advantage of, maybe dfhist?<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Is the dhdl automatically calculated? Where is it stored? If dhdl isn?t<br>
&gt;&gt;&gt; being calculated automatically, is there an mdp option or mdrun flag<br>
&gt;&gt;&gt; that<br>
&gt;&gt;&gt; I need?<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; I appreciate any advice that can be offered.<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; -ChrisM<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt; Mirabzadeh, Christopher<br>
&gt;&gt;&gt; Graduate Research Assistant/Physics Instructor<br>
&gt;&gt;&gt; Department of Physics<br>
&gt;&gt;&gt; University of Idaho<br>
&gt;&gt;&gt; Moscow, Id<br>
&gt;&gt;&gt; <a href="tel:%28509%29339-5647" value="+15093395647">(509)339-5647</a><br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;&gt;<br>
&gt;&gt;<br>
&gt;<br>
&gt; --<br>
&gt; Gromacs Developers mailing list<br>
&gt;<br>
&gt; * Please search the archive at <a href="http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List</a> before posting!<br>
&gt;<br>
&gt; * Can&#39;t post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
&gt;<br>
&gt; * For (un)subscribe requests visit<br>
&gt; <a href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers" rel="noreferrer" target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers</a> or send a mail to <a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a>.<br>
<br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Fri, 28 Aug 2015 18:38:25 +0200<br>
From: Berk Hess &lt;<a href="mailto:hess@kth.se">hess@kth.se</a>&gt;<br>
To: <a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
Subject: Re: [gmx-developers] Increase the scope of dd domains<br>
Message-ID: &lt;<a href="mailto:55E08E81.2020207@kth.se">55E08E81.2020207@kth.se</a>&gt;<br>
Content-Type: text/plain; charset=&quot;windows-1252&quot;; Format=&quot;flowed&quot;<br>
<br>
Hi,<br>
<br>
I don&#39;t see why your current code doesn&#39;t work with 1D DD.<br>
<br>
You do call this code after the coordinate communication has been done?<br>
<br>
Cheers,<br>
<br>
Berk<br>
<br>
On 08/28/2015 05:45 PM, Xingcheng Lin wrote:<br>
&gt; Hi Berk,<br>
&gt;<br>
&gt; The 1D dd where I used 2 MPI processes with -dd 1 1 2 can only work if<br>
&gt; I do not turn on the -dlb yes. It seems the problem will always exist<br>
&gt; if I have atom pairs separated in two domains.<br>
&gt;<br>
&gt; You mentioned I could get all pairs by only looking at home atoms for<br>
&gt; the i-atoms. Does it mean I can even make the code like this, by<br>
&gt; changing ga2la_get for aj to ga2la_get_home?<br>
&gt;<br>
&gt; // NP is number of tested pairs recorded in an array;<br>
&gt; // If atom ai is one of the home atoms in a node, we start testing if<br>
&gt; aj is<br>
&gt; available by the same node;<br>
&gt; // rlist = 1.0nm, Coulombtype  =  Cut-off, cutoff-scheme = Verlet, ns_type<br>
&gt; = grid, pbc = xyz;<br>
&gt;<br>
&gt; int Npairs = 0;<br>
&gt; for (i=0; i&lt;NP; ++i)<br>
&gt; {<br>
&gt;      int ai = datai[i] ;<br>
&gt;      int localai;<br>
&gt;      // Examine if ai it is within the home atoms index;<br>
&gt;      if (ga2la_get_home(cr-&gt;dd-&gt;ga2la, ai, &amp;localai))<br>
&gt;      {<br>
&gt;         aj = dataj[i] ;<br>
&gt;         // Examine if aj is available by local node;;<br>
&gt;         if (ga2la_get_home(cr-&gt;dd-&gt;ga2la, aj, &amp;localaj))<br>
&gt;         {<br>
&gt;            FILE *fp;<br>
&gt;            fp = fopen(&quot;pairs.dat&quot;,&quot;a&quot;);<br>
&gt;<br>
&gt;            fprintf(fp, &quot;%d\t%d\n&quot;, ai, aj);<br>
&gt;            fclose(fp);<br>
&gt;<br>
&gt;            r_N = 0.6 ;<br>
&gt;            pbc_dx(&amp;pbc, x[localai], x[localaj], dx) ;<br>
&gt;            dr2 = iprod( dx, dx ) ;<br>
&gt;            dr  = sqrt( dr2 ) ;<br>
&gt;            if (dr &lt; r_N)<br>
&gt;            {<br>
&gt;                Npairs += 1;<br>
&gt;            }<br>
&gt;         }<br>
&gt;      }<br>
&gt;  }<br>
&gt;<br>
&gt;<br>
&gt; On Fri, Aug 28, 2015 at 2:22 AM,<br>
&gt; &lt;<a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a><br>
&gt; &lt;mailto:<a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a>&gt;&gt; wrote:<br>
&gt;<br>
&gt;     Send gromacs.org_gmx-developers mailing list submissions to<br>
&gt;     <a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&gt;<br>
&gt;<br>
&gt;     To subscribe or unsubscribe via the World Wide Web, visit<br>
&gt;     <a href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers" rel="noreferrer" target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers</a><br>
&gt;<br>
&gt;     or, via email, send a message with subject or body &#39;help&#39; to<br>
&gt;     <a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers-request@maillist.sys.kth.se">gromacs.org_gmx-developers-request@maillist.sys.kth.se</a>&gt;<br>
&gt;<br>
&gt;     You can reach the person managing the list at<br>
&gt;     <a href="mailto:gromacs.org_gmx-developers-owner@maillist.sys.kth.se">gromacs.org_gmx-developers-owner@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers-owner@maillist.sys.kth.se">gromacs.org_gmx-developers-owner@maillist.sys.kth.se</a>&gt;<br>
&gt;<br>
&gt;     When replying, please edit your Subject line so it is more specific<br>
&gt;     than &quot;Re: Contents of gromacs.org_gmx-developers digest...&quot;<br>
&gt;<br>
&gt;<br>
&gt;     Today&#39;s Topics:<br>
&gt;<br>
&gt;        1. Increase the scope of dd domains (Xingcheng Lin)<br>
&gt;        2. Re: energy group computation (Berk Hess)<br>
&gt;        3. Re: dhdl expanded ensemble (Berk Hess)<br>
&gt;        4. Re: Increase the scope of dd domains (Berk Hess)<br>
&gt;<br>
&gt;<br>
&gt;     ----------------------------------------------------------------------<br>
&gt;<br>
&gt;     Message: 1<br>
&gt;     Date: Thu, 27 Aug 2015 23:35:30 -0500<br>
&gt;     From: Xingcheng Lin &lt;<a href="mailto:linxingcheng50311@gmail.com">linxingcheng50311@gmail.com</a><br>
&gt;     &lt;mailto:<a href="mailto:linxingcheng50311@gmail.com">linxingcheng50311@gmail.com</a>&gt;&gt;<br>
&gt;     To: <a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&gt;<br>
&gt;     Subject: [gmx-developers] Increase the scope of dd domains<br>
&gt;     Message-ID:<br>
&gt;<br>
&gt;     &lt;CAPugp-HQjXb8hYKJO9i-1fMSfpzQxe=<a href="mailto:qZCOo8tPwPQrFraKEmw@mail.gmail.com">qZCOo8tPwPQrFraKEmw@mail.gmail.com</a> &lt;mailto:<a href="mailto:qZCOo8tPwPQrFraKEmw@mail.gmail.com">qZCOo8tPwPQrFraKEmw@mail.gmail.com</a>&gt;&gt;<br>
&gt;     Content-Type: text/plain; charset=&quot;utf-8&quot;<br>
&gt;<br>
&gt;     Hi,<br>
&gt;<br>
&gt;     I am trying to encode a &quot;on the fly&quot; calculation for the number of<br>
&gt;     atom<br>
&gt;     pairs within some distance (for example here, 6 angstroms).<br>
&gt;<br>
&gt;     The problem happens when doing parallel domain decomposition.<br>
&gt;     Specifically,<br>
&gt;     when two particles reside in different domains, gromacs is not able to<br>
&gt;     recognize this pair. I am trying to use ga2la_get_home and<br>
&gt;     ga2la_get for<br>
&gt;     calculation, here is the code:<br>
&gt;<br>
&gt;     // NP is number of tested pairs recorded in an array;<br>
&gt;     // If atom ai is one of the home atoms in a node, we start testing<br>
&gt;     if aj is<br>
&gt;     available by the same node;<br>
&gt;     // rlist = 1.0nm, Coulombtype  =  Cut-off, cutoff-scheme = Verlet,<br>
&gt;     ns_type<br>
&gt;     = grid, pbc = xyz;<br>
&gt;<br>
&gt;     int Npairs = 0;<br>
&gt;     for (i=0; i&lt;NP; ++i)<br>
&gt;     {<br>
&gt;          int ai = datai[i] ;<br>
&gt;          int localai;<br>
&gt;          // Examine if ai it is within the home atoms index;<br>
&gt;          if (ga2la_get_home(cr-&gt;dd-&gt;ga2la, ai, &amp;localai))<br>
&gt;          {<br>
&gt;             aj = dataj[i] ;<br>
&gt;             // Examine if aj is available by local node;;<br>
&gt;             if (ga2la_get(cr-&gt;dd-&gt;ga2la, aj, &amp;localaj, &amp;localcell))<br>
&gt;             {<br>
&gt;                FILE *fp;<br>
&gt;                fp = fopen(&quot;pairs.dat&quot;,&quot;a&quot;);<br>
&gt;<br>
&gt;                fprintf(fp, &quot;%d\t%d\n&quot;, ai, aj);<br>
&gt;                fclose(fp);<br>
&gt;<br>
&gt;                r_N = 0.6 ;<br>
&gt;                pbc_dx(&amp;pbc, x[localai], x[localaj], dx) ;<br>
&gt;                dr2 = iprod( dx, dx ) ;<br>
&gt;                dr  = sqrt( dr2 ) ;<br>
&gt;                if (dr &lt; r_N)<br>
&gt;                {<br>
&gt;                    Npairs += 1;<br>
&gt;                }<br>
&gt;             }<br>
&gt;          }<br>
&gt;      }<br>
&gt;<br>
&gt;     The results show some atom pairs, while atoms less than 6<br>
&gt;     angstroms apart,<br>
&gt;     are missing here. I increase rlist, but it doesn&#39;t help. Is there<br>
&gt;     any way<br>
&gt;     to increase the scope of atoms available in each domain so that<br>
&gt;     the missing<br>
&gt;     atoms can be detected?<br>
&gt;<br>
&gt;     Or are there any mistakes out of my notice in this code?<br>
&gt;<br>
&gt;     Thank you,<br>
&gt;     Xingcheng<br>
&gt;     -------------- next part --------------<br>
&gt;     An HTML attachment was scrubbed...<br>
&gt;     URL:<br>
&gt;     &lt;<a href="http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150827/208ffc05/attachment.html" rel="noreferrer" target="_blank">http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150827/208ffc05/attachment.html</a>&gt;<br>
&gt;<br>
&gt;     ------------------------------<br>
&gt;<br>
&gt;     Message: 2<br>
&gt;     Date: Fri, 28 Aug 2015 09:04:56 +0200<br>
&gt;     From: Berk Hess &lt;<a href="mailto:hess@kth.se">hess@kth.se</a> &lt;mailto:<a href="mailto:hess@kth.se">hess@kth.se</a>&gt;&gt;<br>
&gt;     To: <a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a> &lt;mailto:<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;<br>
&gt;     Subject: Re: [gmx-developers] energy group computation<br>
&gt;     Message-ID: &lt;<a href="mailto:55E00818.1060601@kth.se">55E00818.1060601@kth.se</a> &lt;mailto:<a href="mailto:55E00818.1060601@kth.se">55E00818.1060601@kth.se</a>&gt;&gt;<br>
&gt;     Content-Type: text/plain; charset=&quot;windows-1252&quot;; Format=&quot;flowed&quot;<br>
&gt;<br>
&gt;     Hi,<br>
&gt;<br>
&gt;     This is a use, not a development question. Please post such<br>
&gt;     questions to<br>
&gt;     gmx-users.<br>
&gt;<br>
&gt;     But the quick answer is that although it&#39;s technically possible to<br>
&gt;     calculate this number, this number is useless (and indeed high).<br>
&gt;<br>
&gt;     Cheers,<br>
&gt;<br>
&gt;     Berk<br>
&gt;<br>
&gt;     On 2015-08-27 19:00, ANTHONY C MANSON wrote:<br>
&gt;     &gt; Hello:<br>
&gt;     &gt;<br>
&gt;     &gt; I am trying to estimate the polar solvation energy of a 76 residue<br>
&gt;     &gt; protein at standard conditions.  I am using the gromacs 4.5.3 dp<br>
&gt;     release.<br>
&gt;     &gt; I have set up a simulation with the solvated (using spce water)<br>
&gt;     protein.<br>
&gt;     &gt; I have defined two energygrps: Protein SOL in the *.mdp file.<br>
&gt;     &gt; I have run the simulation and have then run g_energy on the<br>
&gt;     *.edr file.<br>
&gt;     &gt; I get the following choices from g_energy:<br>
&gt;     &gt;<br>
&gt;     &gt;   1  Bond       2  Angle            3 Proper-Dih.      4<br>
&gt;     Improper-Dih.<br>
&gt;     &gt;   5  LJ-14      6  Coulomb-14       7  LJ-(SR)         8<br>
&gt;     Disper.-corr.<br>
&gt;     &gt; 9  Coulomb-(SR)    10  Coul.-recip.    11 Potential       12<br>
&gt;     &gt;  Kinetic-En.<br>
&gt;     &gt;  13  Total-Energy    14  Temperature     15 Pres.-DC        16<br>
&gt;     Pressure<br>
&gt;     &gt;  17  Vir-XX        18  Vir-XY          19 Vir-XZ          20  Vir-YX<br>
&gt;     &gt;  21  Vir-YY  22  Vir-YZ          23  Vir-ZX     24  Vir-ZY<br>
&gt;     &gt;  25  Vir-ZZ          26  Pres-XX         27 Pres-XY   28  Pres-XZ<br>
&gt;     &gt;  29  Pres-YX         30  Pres-YY   31  Pres-YZ        32  Pres-ZX<br>
&gt;     &gt;  33  Pres-ZY   34  Pres-ZZ         35 #Surf*SurfTen   36  Mu-X<br>
&gt;     &gt;  37  Mu-Y              38  Mu-Z<br>
&gt;     &gt;  39  Coul-SR:Protein-Protein         40 LJ-SR:Protein-Protein<br>
&gt;     &gt;  41  Coul-14:Protein-Protein             42 LJ-14:Protein-Protein<br>
&gt;     &gt;  43  Coul-SR:Protein-SOL   44 LJ-SR:Protein-SOL<br>
&gt;     &gt;  45  Coul-14:Protein-SOL             46 LJ-14:Protein-SOL<br>
&gt;     &gt;  47  Coul-SR:SOL-SOL     48  LJ-SR:SOL-SOL<br>
&gt;     &gt;  49  Coul-14:SOL-SOL                     50 LJ-14:SOL-SOL<br>
&gt;     &gt;  51  T-System<br>
&gt;     &gt;<br>
&gt;     &gt; I pick Coul-SR:Protein-SOL as a component of the polar solvation<br>
&gt;     energy.<br>
&gt;     &gt; I note that the energy magnitude seems extremely high:<br>
&gt;     &gt;<br>
&gt;     &gt; Energy              Average   Err.Est.  RMSD  Tot-Drift<br>
&gt;     &gt;<br>
&gt;     -------------------------------------------------------------------------------<br>
&gt;     &gt; Coul-SR:Protein-SOL        -8343.86         46   193.457  -192.93<br>
&gt;     &gt;  (kJ/mol)<br>
&gt;     &gt;<br>
&gt;     &gt; This seems unreasonable to me.  I wanted to understand how mdrun<br>
&gt;     &gt; was calculating this value.  Perhaps it is accumulating energies<br>
&gt;     over<br>
&gt;     &gt; many frames.  What part of the code (force.c ...) handles this<br>
&gt;     &gt; calculation so I can verify its action.<br>
&gt;     &gt;<br>
&gt;     &gt; Also, I tried running g_enemat on this with the following<br>
&gt;     groups.dat file.<br>
&gt;     &gt;<br>
&gt;     &gt; 2<br>
&gt;     &gt; Protein<br>
&gt;     &gt; SOL<br>
&gt;     &gt;<br>
&gt;     &gt; I use the command:<br>
&gt;     &gt;<br>
&gt;     &gt; g_enemat  -f ub_short.edr -e 100<br>
&gt;     &gt;<br>
&gt;     &gt; and get:<br>
&gt;     &gt;<br>
&gt;     &gt; Opened ub_short.edr as double precision energy file<br>
&gt;     &gt; Will read groupnames from in! putfile<br>
&gt;     &gt; Read 2 groups<br>
&gt;     &gt; group 0WARNING! could not find group (null):Protein-Protein (0,0)in<br>
&gt;     &gt; energy file<br>
&gt;     &gt; WARNING! could not find group (null):Protein-SOL (0,1)in energy file<br>
&gt;     &gt; group 1WARNING! could not find group (null):SOL-SOL (1,1)in<br>
&gt;     energy file<br>
&gt;     &gt;<br>
&gt;     &gt; Will select half-matrix of energies with 6 elements<br>
&gt;     &gt; Read frame: 1000, Time: 100.000e  100.000<br>
&gt;     &gt; Will build energy half-matrix of 2 groups, 6 elements, over 1001<br>
&gt;     frames<br>
&gt;     &gt; Segmentation fault<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt; Cheers<br>
&gt;     &gt; Anthony Manson Ph.D.<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;<br>
&gt;     -------------- next part --------------<br>
&gt;     An HTML attachment was scrubbed...<br>
&gt;     URL:<br>
&gt;     &lt;<a href="http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/17c227f6/attachment-0001.html" rel="noreferrer" target="_blank">http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/17c227f6/attachment-0001.html</a>&gt;<br>
&gt;<br>
&gt;     ------------------------------<br>
&gt;<br>
&gt;     Message: 3<br>
&gt;     Date: Fri, 28 Aug 2015 09:09:49 +0200<br>
&gt;     From: Berk Hess &lt;<a href="mailto:hess@kth.se">hess@kth.se</a> &lt;mailto:<a href="mailto:hess@kth.se">hess@kth.se</a>&gt;&gt;<br>
&gt;     To: <a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
&gt;     &lt;mailto:<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;, &quot;<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a><br>
&gt;     &lt;mailto:<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a>&gt;&quot;<br>
&gt;             &lt;<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a><br>
&gt;     &lt;mailto:<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a>&gt;&gt;<br>
&gt;     Subject: Re: [gmx-developers] dhdl expanded ensemble<br>
&gt;     Message-ID: &lt;<a href="mailto:55E0093D.2000804@kth.se">55E0093D.2000804@kth.se</a> &lt;mailto:<a href="mailto:55E0093D.2000804@kth.se">55E0093D.2000804@kth.se</a>&gt;&gt;<br>
&gt;     Content-Type: text/plain; charset=windows-1252; format=flowed<br>
&gt;<br>
&gt;     No. To be more precise dhdl is only (fully) calculated at each step<br>
&gt;     where free energies are calculated, i.e. every nstdhdl steps.<br>
&gt;     To find where the final values are stored, it&#39;s often easiest to<br>
&gt;     look in<br>
&gt;     mdebin.c, in this case at line 1244.<br>
&gt;<br>
&gt;     Cheers,<br>
&gt;<br>
&gt;     Berk<br>
&gt;<br>
&gt;     On 2015-08-28 03:52, Shirts, Michael R. (mrs5pt) wrote:<br>
&gt;     &gt; Dhdl is automatically calculated at every step if free energies are<br>
&gt;     &gt; calculated. Look in sum_dhdl -- difference components are stored in<br>
&gt;     &gt; enerd-&gt;term.<br>
&gt;     &gt;<br>
&gt;     &gt; Unless it&#39;s a new development you would rather keep private for<br>
&gt;     now, you<br>
&gt;     &gt; should consider posting the proposed enhancement on redmine.<br>
&gt;     You would<br>
&gt;     &gt; likely get some tips about how to do it in a way that could be<br>
&gt;     more easily<br>
&gt;     &gt; included in later gromcs versions.<br>
&gt;     &gt;<br>
&gt;     &gt; Best,<br>
&gt;     &gt; ~~~~~~~~~~~~<br>
&gt;     &gt; Michael Shirts<br>
&gt;     &gt; Associate Professor<br>
&gt;     &gt; Department of Chemical Engineering<br>
&gt;     &gt; University of Virginia<br>
&gt;     &gt; <a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a> &lt;mailto:<a href="mailto:michael.shirts@virginia.edu">michael.shirts@virginia.edu</a>&gt;<br>
&gt;     &gt; <a href="tel:%28434%29%20243-1821" value="+14342431821">(434) 243-1821</a> &lt;tel:%28434%29%20243-1821&gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt; From:  &lt;Mirabzadeh&gt;, &quot;Christopher   (<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a><br>
&gt;     &lt;mailto:<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>&gt;)&quot;<br>
&gt;     &gt; &lt;<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a> &lt;mailto:<a href="mailto:mira2978@vandals.uidaho.edu">mira2978@vandals.uidaho.edu</a>&gt;&gt;<br>
&gt;     &gt; Reply-To:  &quot;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
&gt;     &lt;mailto:<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;&quot; &lt;<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a><br>
&gt;     &lt;mailto:<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;&gt;<br>
&gt;     &gt; Date:  Thursday, August 27, 2015 at 12:20 PM<br>
&gt;     &gt; To:  &quot;<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&gt;&quot;<br>
&gt;     &gt; &lt;<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a><br>
&gt;     &lt;mailto:<a href="mailto:gromacs.org_gmx-developers@maillist.sys.kth.se">gromacs.org_gmx-developers@maillist.sys.kth.se</a>&gt;&gt;<br>
&gt;     &gt; Subject:  [gmx-developers]  dhdl expanded ensemble<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt; Hello,<br>
&gt;     &gt;<br>
&gt;     &gt; What I&#39;m trying to do;<br>
&gt;     &gt;<br>
&gt;     &gt; I&#39;m attempting to add functionality to expanded ensemble.<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt; What I&#39;ve done so far;<br>
&gt;     &gt;<br>
&gt;     &gt; I&#39;ve added functionality to accept my mdp options and found<br>
&gt;     pre-written<br>
&gt;     &gt; functions to take advantage of.<br>
&gt;     &gt;<br>
&gt;     &gt; Where I&#39;m stuck;<br>
&gt;     &gt;<br>
&gt;     &gt; I need the derivative of the Hamiltonian with respect to lambda<br>
&gt;     at every<br>
&gt;     &gt; md step. I&#39;m having a hard time narrowing down this variable. I<br>
&gt;     know it&#39;s<br>
&gt;     &gt; calculated and I see in mdebin.c there is a line that says &quot;/*<br>
&gt;     store_dh is<br>
&gt;     &gt; dE */&quot;. Is the derivative &quot;store_dhdl[]=enerd-&gt;term[F_DVDL]&quot;?<br>
&gt;     &gt;    Or, is this already being called somewhere in expanded.c that<br>
&gt;     I can take<br>
&gt;     &gt; advantage of, maybe dfhist?<br>
&gt;     &gt;<br>
&gt;     &gt; Is the dhdl automatically calculated? Where is it stored? If<br>
&gt;     dhdl isn?t<br>
&gt;     &gt; being calculated automatically, is there an mdp option or mdrun<br>
&gt;     flag that<br>
&gt;     &gt; I need?<br>
&gt;     &gt;<br>
&gt;     &gt; I appreciate any advice that can be offered.<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt; -ChrisM<br>
&gt;     &gt;<br>
&gt;     &gt; Mirabzadeh, Christopher<br>
&gt;     &gt; Graduate Research Assistant/Physics Instructor<br>
&gt;     &gt; Department of Physics<br>
&gt;     &gt; University of Idaho<br>
&gt;     &gt; Moscow, Id<br>
&gt;     &gt; <a href="tel:%28509%29339-5647" value="+15093395647">(509)339-5647</a> &lt;tel:%28509%29339-5647&gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;     ------------------------------<br>
&gt;<br>
&gt;     Message: 4<br>
&gt;     Date: Fri, 28 Aug 2015 09:20:44 +0200<br>
&gt;     From: Berk Hess &lt;<a href="mailto:hess@kth.se">hess@kth.se</a> &lt;mailto:<a href="mailto:hess@kth.se">hess@kth.se</a>&gt;&gt;<br>
&gt;     To: <a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a> &lt;mailto:<a href="mailto:gmx-developers@gromacs.org">gmx-developers@gromacs.org</a>&gt;<br>
&gt;     Subject: Re: [gmx-developers] Increase the scope of dd domains<br>
&gt;     Message-ID: &lt;<a href="mailto:55E00BCC.4040506@kth.se">55E00BCC.4040506@kth.se</a> &lt;mailto:<a href="mailto:55E00BCC.4040506@kth.se">55E00BCC.4040506@kth.se</a>&gt;&gt;<br>
&gt;     Content-Type: text/plain; charset=&quot;windows-1252&quot;; Format=&quot;flowed&quot;<br>
&gt;<br>
&gt;     Hi,<br>
&gt;<br>
&gt;     Gromacs uses a neutral-territory eighth-shell domain decomposition<br>
&gt;     method. So you can get all pairs by only looking at home-atoms for the<br>
&gt;     i-atoms. You code should work though for 1D decomposition, is that<br>
&gt;     the case?<br>
&gt;<br>
&gt;     To get this to work with 2D or 3D decomposition, you need to replace<br>
&gt;     ga2la_get_home by ga2la_get and then only consider i-j pairs where the<br>
&gt;     cell values are in the cell pair list given by dd-&gt;comm-&gt;zones. cell_i<br>
&gt;     should be &lt; zones-&gt;nizone and zones-&gt;izone[cell_i].j0 &lt;= cell_j &lt;<br>
&gt;     zones-&gt;izone[cell_j].j1 When the DD cells get smaller than the cutoff,<br>
&gt;     you need to use cell % zones-&gt;n.<br>
&gt;<br>
&gt;     Note that since the ga2la_get calls are expensive, such an approach is<br>
&gt;     only efficient when you have only few atoms / atom pairs.<br>
&gt;     For improving performance, it can help to first lookup locala and cell<br>
&gt;     for all atoms and store them and then do the pair double loop after<br>
&gt;     that. And instead of using sqrt, check dr2 &lt; r_N*r_N.<br>
&gt;<br>
&gt;     Cheers,<br>
&gt;<br>
&gt;     Berk<br>
&gt;<br>
&gt;     On 2015-08-28 06:35, Xingcheng Lin wrote:<br>
&gt;     &gt; Hi,<br>
&gt;     &gt;<br>
&gt;     &gt; I am trying to encode a &quot;on the fly&quot; calculation for the number of<br>
&gt;     &gt; atom pairs within some distance (for example here, 6 angstroms).<br>
&gt;     &gt;<br>
&gt;     &gt; The problem happens when doing parallel domain decomposition.<br>
&gt;     &gt; Specifically, when two particles reside in different domains,<br>
&gt;     gromacs<br>
&gt;     &gt; is not able to recognize this pair. I am trying to use<br>
&gt;     ga2la_get_home<br>
&gt;     &gt; and ga2la_get for calculation, here is the code:<br>
&gt;     &gt;<br>
&gt;     &gt; // NP is number of tested pairs recorded in an array;<br>
&gt;     &gt; // If atom ai is one of the home atoms in a node, we start<br>
&gt;     testing if<br>
&gt;     &gt; aj is available by the same node;<br>
&gt;     &gt; // rlist = 1.0nm, Coulombtype  =  Cut-off, cutoff-scheme = Verlet,<br>
&gt;     &gt; ns_type  = grid, pbc = xyz;<br>
&gt;     &gt;<br>
&gt;     &gt; int Npairs = 0;<br>
&gt;     &gt; for (i=0; i&lt;NP; ++i)<br>
&gt;     &gt; {<br>
&gt;     &gt;      int ai = datai[i] ;<br>
&gt;     &gt;      int localai;<br>
&gt;     &gt;      // Examine if ai it is within the home atoms index;<br>
&gt;     &gt;      if (ga2la_get_home(cr-&gt;dd-&gt;ga2la, ai, &amp;localai))<br>
&gt;     &gt;      {<br>
&gt;     &gt;         aj = dataj[i] ;<br>
&gt;     &gt;         // Examine if aj is available by local node;;<br>
&gt;     &gt;         if (ga2la_get(cr-&gt;dd-&gt;ga2la, aj, &amp;localaj, &amp;localcell))<br>
&gt;     &gt;         {<br>
&gt;     &gt;            FILE *fp;<br>
&gt;     &gt;            fp = fopen(&quot;pairs.dat&quot;,&quot;a&quot;);<br>
&gt;     &gt;<br>
&gt;     &gt;            fprintf(fp, &quot;%d\t%d\n&quot;, ai, aj);<br>
&gt;     &gt;            fclose(fp);<br>
&gt;     &gt;<br>
&gt;     &gt;            r_N = 0.6 ;<br>
&gt;     &gt;            pbc_dx(&amp;pbc, x[localai], x[localaj], dx) ;<br>
&gt;     &gt;            dr2 = iprod( dx, dx ) ;<br>
&gt;     &gt;            dr  = sqrt( dr2 ) ;<br>
&gt;     &gt;            if (dr &lt; r_N)<br>
&gt;     &gt;            {<br>
&gt;     &gt;                Npairs += 1;<br>
&gt;     &gt;            }<br>
&gt;     &gt;         }<br>
&gt;     &gt;      }<br>
&gt;     &gt;  }<br>
&gt;     &gt;<br>
&gt;     &gt; The results show some atom pairs, while atoms less than 6 angstroms<br>
&gt;     &gt; apart, are missing here. I increase rlist, but it doesn&#39;t help. Is<br>
&gt;     &gt; there any way to increase the scope of atoms available in each<br>
&gt;     domain<br>
&gt;     &gt; so that the missing atoms can be detected?<br>
&gt;     &gt;<br>
&gt;     &gt; Or are there any mistakes out of my notice in this code?<br>
&gt;     &gt;<br>
&gt;     &gt; Thank you,<br>
&gt;     &gt; Xingcheng<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;     &gt;<br>
&gt;<br>
&gt;     -------------- next part --------------<br>
&gt;     An HTML attachment was scrubbed...<br>
&gt;     URL:<br>
&gt;     &lt;<a href="http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/682fbb3e/attachment.html" rel="noreferrer" target="_blank">http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/682fbb3e/attachment.html</a>&gt;<br>
&gt;<br>
&gt;     ------------------------------<br>
&gt;<br>
&gt;     --<br>
&gt;     Gromacs Developers mailing list<br>
&gt;<br>
&gt;     * Please search the archive at<br>
&gt;     <a href="http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List</a><br>
&gt;     before posting!<br>
&gt;<br>
&gt;     * Can&#39;t post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
&gt;<br>
&gt;     * For (un)subscribe requests visit<br>
&gt;     <a href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers" rel="noreferrer" target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers</a><br>
&gt;     or send a mail to <a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a><br>
&gt;     &lt;mailto:<a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a>&gt;.<br>
&gt;<br>
&gt;     End of gromacs.org_gmx-developers Digest, Vol 136, Issue 21<br>
&gt;     ***********************************************************<br>
&gt;<br>
&gt;<br>
&gt;<br>
&gt;<br>
<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: &lt;<a href="http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/9bbc372e/attachment.html" rel="noreferrer" target="_blank">http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-developers/attachments/20150828/9bbc372e/attachment.html</a>&gt;<br>
<br>
------------------------------<br>
<span class=""><font color="#888888"><br>
--<br>
Gromacs Developers mailing list<br>
<br>
* Please search the archive at <a href="http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists/GMX-developers_List</a> before posting!<br>
<br>
* Can&#39;t post? Read <a href="http://www.gromacs.org/Support/Mailing_Lists" rel="noreferrer" target="_blank">http://www.gromacs.org/Support/Mailing_Lists</a><br>
<br>
* For (un)subscribe requests visit<br>
<a href="https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers" rel="noreferrer" target="_blank">https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-developers</a> or send a mail to <a href="mailto:gmx-developers-request@gromacs.org">gmx-developers-request@gromacs.org</a>.<br>
<br>
End of gromacs.org_gmx-developers Digest, Vol 136, Issue 23<br>
***********************************************************<br>
</font></span></blockquote></div><br></div></div></div></div></div></div></div></div></div></div>