Actually, I have intention to combine both search strategies: dihedral and cartezian.<br><br>I had developed my main loop and I committed it in my repository [1]. Please, look for do_peo function.<br>However, I'm developing the dihedral computation. <br>
<br>By the way, the pseudo code of my main loop can be see below:<br><br>while(generation<ngenerations)<div>{<br> if (generation > 0){<br>
bRerun = TRUE;<br> } <br> while(step<nsteps){<br>
dd_partition_system(): redistribute atoms every neighbor search steps, calls atoms2md() with index<br>
do_force(): calculate forces (bonded and nonbonded) acting on the (home) atoms (no position update here)<br> write_traj(): write current coordinates to trajectory file (gathers coordinates before hands if MPI-enabled)<br>
update(): tcoupl, constraints, second part of leap frog, update X coords<br> step++;<br> }<br> //Call Evolutionary Algorithm (NSGA-II) to build new conformation based on the energy of conformations. Only here, we'll work with dihedral angles. I'm working here. <br>
buildnewconformation();<br> //According fitness values choose the best solution (conformation).<br>
write_traj(): write current coordinates to trajectory file from best solution<br> generation++<br>}<br><br></div>Thanks in advance,<br><br>[1] <a href="http://gitorious.org/protpred-peo/gromacs/blobs/master/src/kernel/md.c">http://gitorious.org/protpred-peo/gromacs/blobs/master/src/kernel/md.c</a><br clear="all">
--<br>Rodrigo Antonio Faccioli<br>Ph.D Student in Electrical Engineering<br>University of Sao Paulo - USP<br>
Engineering School of Sao Carlos - EESC<br>Department of Electrical Engineering - SEL<br>Intelligent System in Structure Bioinformatics<br><a href="http://laips.sel.eesc.usp.br" target="_blank">http://laips.sel.eesc.usp.br</a><br>
Phone: 55 (16) 3373-9366 Ext 229<br>
Curriculum Lattes - <a href="http://lattes.cnpq.br/1025157978990218" target="_blank">http://lattes.cnpq.br/1025157978990218</a><br>Public Profile - <a href="http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5" target="_blank">http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5</a><br>
<br><br><div class="gmail_quote">On Fri, Jun 18, 2010 at 3:33 PM, David van der Spoel <span dir="ltr"><<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a>></span> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
<div>On 2010-06-18 20.12, Rodrigo Faccioli wrote:<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
David, thanks for your answer.<br>
<br>
I would like to know more details about your ideas to analyse cartesian<br>
space rather than dihedral space.<br>
<br>
I understand that sampling in dihedral space may contain some problems.<br>
However, it allows to exploit the search space with a lower computation<br>
time.<br>
<br>
The dihedral representation is common in Evolutionary Algorithms (EA)<br>
based on not only in its power of exploit the search space but also, in<br>
thesis, can select the best conformations because the worse<br>
conformations (superposition of atoms, for example) have greater energy<br>
value. So, these conformations have less change to be selected for new<br>
generation.<br>
<br>
</blockquote></div>
Exactly, it may be easier to generate non-sensical conformations using dihedral angle confs. but they may be filtered out quickly as well. In cartesian space one could do a lot of small steps that have no effect. What is better in the end I'm not sure. You will have to test...<br>
<br>
Programs like autodock also use a genetic algorithm in dihedral space, and are deemed to be efficient, as long as the number of degrees of freedom is low.<br>
<br>
<blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;"><div>
Sorry if my explanation about dihedral representation in EA isn't<br>
enough. I'm computer scientist try understanding about some physical<br>
properties. I might make mistakes :-)<br>
<br>
Thanks in advance,<br>
<br>
--<br>
Rodrigo Antonio Faccioli<br>
Ph.D Student in Electrical Engineering<br>
University of Sao Paulo - USP<br>
Engineering School of Sao Carlos - EESC<br>
Department of Electrical Engineering - SEL<br>
Intelligent System in Structure Bioinformatics<br>
<a href="http://laips.sel.eesc.usp.br" target="_blank">http://laips.sel.eesc.usp.br</a><br>
Phone: 55 (16) 3373-9366 Ext 229<br>
Curriculum Lattes - <a href="http://lattes.cnpq.br/1025157978990218" target="_blank">http://lattes.cnpq.br/1025157978990218</a><br>
Public Profile - <a href="http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5" target="_blank">http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5</a><br>
<br>
<br>
On Wed, Jun 16, 2010 at 6:08 PM, David van der Spoel<br></div><div><div></div><div>
<<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a> <mailto:<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a>>> wrote:<br>
<br>
On 6/16/10 8:04 PM, Rodrigo Faccioli wrote:<br>
<br>
Hi,<br>
<br>
We've been working with ProtPred-PEO-Gromacs project [1]. In general<br>
lines, this project is a implementation of NSGA-II algorithm in<br>
Gromacs<br>
to discover a minima energy of protein conformation . NSGA-II is a<br>
Multi-Objective Evolutionary Algorithm (MOEA).<br>
<br>
More specific in Gromacs implementation, this project is a<br>
multi-system<br>
as REMD simulation. However, the new protein conformations are<br>
obtained<br>
from genetic operators. In [2] there is a simple documentation<br>
wrote by<br>
me to publish the some project ideas about it.<br>
<br>
The integrator option is peo instead of md and its function is<br>
do_peo in<br>
md.c. This project is under development version. My genetic<br>
operators<br>
work with the protein backbone angles. This representation is<br>
common for<br>
MOEAs when is working with proteins.<br>
<br>
In this way, I've studied about g_dhi program. I found the dih_angle<br>
function in bondfree.c. This function computes all backbone angles<br>
storing them in t_xrama structure represented by xr variable.<br>
<br>
Because of my simulation is a multi-system, I have for each system<br>
(protein conformation) a trr file. I need ideas to help me to<br>
obtain my<br>
backbone angles without call g_dhi routines..<br>
<br>
I thought to compute the backbone angles in my integration function<br>
(do_peo) because it has all atoms. So, I want to call a function<br>
which<br>
computes the backbone angle of my protein such as:<br>
t_peo_phipsi compute_backbone(t_mdatoms *mdatoms).<br>
<br>
Any comments for a good way to do this? Or other ways?<br>
<br>
<br>
Look in src/gmxlib/bondfree.c to see how this is done internally<br>
when computing dihedral forces. You loop over the right structure to<br>
compute the dihedrals.<br>
<br>
But do you really want to do the sampling in dihedral space?<br>
There are many false minima in dihedral space that are forbidden by<br>
van der waals interactions. It might be easier to do it in cartesian<br>
space.<br>
<br>
<br>
<br>
[1] <a href="http://gitorious.org/protpred-peo/gromacs" target="_blank">http://gitorious.org/protpred-peo/gromacs</a><br>
[2]<br>
<a href="https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en" target="_blank">https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en</a><br>
<<a href="https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en" target="_blank">https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en</a>><br>
<<a href="https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en" target="_blank">https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en</a><br>
<<a href="https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en" target="_blank">https://docs.google.com/fileview?id=0ByNUaKmUm2WoNDRlZDkwYmYtNTBhMS00NGUwLWI1MDMtOWEzMjNmY2I4OGQ5&hl=en</a>>><br>
<br>
<br>
Thanks in advance,<br>
<br>
--<br>
Rodrigo Antonio Faccioli<br>
Ph.D Student in Electrical Engineering<br>
University of Sao Paulo - USP<br>
Engineering School of Sao Carlos - EESC<br>
Department of Electrical Engineering - SEL<br>
Intelligent System in Structure Bioinformatics<br>
<a href="http://laips.sel.eesc.usp.br" target="_blank">http://laips.sel.eesc.usp.br</a><br>
Phone: 55 (16) 3373-9366 Ext 229<br>
Curriculum Lattes - <a href="http://lattes.cnpq.br/1025157978990218" target="_blank">http://lattes.cnpq.br/1025157978990218</a><br>
Public Profile -<br>
<a href="http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5" target="_blank">http://br.linkedin.com/pub/rodrigo-faccioli/7/589/a5</a><br>
<br>
<br>
<br>
--<br>
David.<br>
________________________________________________________________________<br>
David van der Spoel, PhD, Professor of Biology<br>
Dept. of Cell and Molecular Biology, Uppsala University.<br>
Husargatan 3, Box 596, 75124 Uppsala, Sweden<br>
phone: 46 18 471 4205 fax: 46 18 511 755<br></div></div>
<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a> <mailto:<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a>> <a href="mailto:spoel@gromacs.org" target="_blank">spoel@gromacs.org</a><br>
<mailto:<a href="mailto:spoel@gromacs.org" target="_blank">spoel@gromacs.org</a>> <a href="http://folding.bmc.uu.se" target="_blank">http://folding.bmc.uu.se</a><br>
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++<br>
--<br>
gmx-developers mailing list<br>
<a href="mailto:gmx-developers@gromacs.org" target="_blank">gmx-developers@gromacs.org</a> <mailto:<a href="mailto:gmx-developers@gromacs.org" target="_blank">gmx-developers@gromacs.org</a>><div><br>
<a href="http://lists.gromacs.org/mailman/listinfo/gmx-developers" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-developers</a><br>
Please don't post (un)subscribe requests to the list. Use the www<br>
interface or send it to <a href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a><br></div>
<mailto:<a href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a>>.<br>
<br>
<br>
</blockquote>
<br>
<br>
-- <br>
David van der Spoel, Ph.D., Professor of Biology<br>
Dept. of Cell & Molec. Biol., Uppsala University.<br>
Box 596, 75124 Uppsala, Sweden. Phone: +46184714205.<br>
<a href="mailto:spoel@xray.bmc.uu.se" target="_blank">spoel@xray.bmc.uu.se</a> <a href="http://folding.bmc.uu.se" target="_blank">http://folding.bmc.uu.se</a><div><div></div><div><br>
-- <br>
gmx-developers mailing list<br>
<a href="mailto:gmx-developers@gromacs.org" target="_blank">gmx-developers@gromacs.org</a><br>
<a href="http://lists.gromacs.org/mailman/listinfo/gmx-developers" target="_blank">http://lists.gromacs.org/mailman/listinfo/gmx-developers</a><br>
Please don't post (un)subscribe requests to the list. Use the www interface or send it to <a href="mailto:gmx-developers-request@gromacs.org" target="_blank">gmx-developers-request@gromacs.org</a>.<br>
</div></div></blockquote></div><br>