[gmx-users] Help on data anaysis in g_rdf and g_dist

Sat Aug 28 02:11:21 CEST 2010

teklebrh at ualberta.ca wrote:
> Dear Justine,
> 
>> As for the distance, have you corrected for PBC? Are you analyzing the 
>> molecules that you think you are?
> 
> No, I did not correct for the PBC. How can I do that. And what is the 
> purpose of that in the g_dist calculation.

Usually one uses trjconv.  I'm guessing a bit here, since the only explanation 
you offered was that something "did not look right."  g_dist is one of those 
reliable tools that performs a very simple calculation, so if you think the 
distance should be 0.5 nm, either the trajectory does not agree with your 
conclusion, or you selected something wrong when choosing your analysis groups.

> 
> Yes,
> 
> I just look at the last frame of my simulation and correctly identified 
> the two residues that form aggregates using the sequence viewers in VMD. 
> and I developed my index file based on these two. I got the residue name 
> and check the group atoms that constitute in the polyaromatic ring of 
> these two molecules and calculated the g_dist for ONLY those two 
> molecules in the last 1ns.
> 
>> By default, RDF is done based on atoms, so all atom-atom distances 
>> will be plotted as part of the RDF. If you want a molecule-based RDF, 
>> see the options for the -rdf flag.
> 
> Yes, I understand that the choise of atom based or molecule based but 
> when you said atom-atom based is that for only two molecules or the 
> entire 24 molecules in the system ( means the two that form aggregates).
> 

If you're choosing simply "PAP" molecules as the group for analysis (with the 
default -rdf atom), then all atom pairs are considered.  Whether or not that's 
what you want is up to you, but g_rdf is very flexible, so do explore your 
options and consider how best to utilize them.

-Justin

> Thank you for you help Justine
> 
> Rob
> 
> Quoting "Justin A. Lemkul" <jalemkul at vt.edu>:
> 
>>
>>
>> teklebrh at ualberta.ca wrote:
>>>
>>> Dear Gmx users,
>>>
>>> After working for quite sometime I managed to simulated almost all my 
>>> molecules for 20ns. Thank you for all your help especially 
>>> Justine.Now I am on last stage of analyzing my data.
>>>
>>> First:
>>> I want to calculate g(r)and I used the following command
>>>
>>> g_rdf -f PAP_20ns.xtc -s PAP_20ns.tpr -n PAP.ndx -b 19000 -e 20000 -o 
>>> result.xvg
>>>
>>> Is this right and my data does not look right. For example for the 
>>> PAP-PAP radial distribution function , My data did not converge to 
>>> one at the end it goes all the way to zero. Which is wierd. when I 
>>> made the index file do I need to make for each molecules in the 
>>> system or I cann collectively get the g(r) for all the PAP molecules 
>>> in the system. PAP-PAP RDF for individual molecues in the index file 
>>> or for two separate g(r) is the right value to trust for g(r) of the 
>>> PAP molecule.
>>>
>>
>> By default, RDF is done based on atoms, so all atom-atom distances 
>> will be plotted as part of the RDF. If you want a molecule-based RDF, 
>> see the options for the -rdf flag.
>>
>>> Second:
>>>
>>> the distance between two polyaromatic rings. I develop a make index 
>>> as follows.
>>>
>>> [System]
>>>
>>> -
>>> [PAP]
>>> -
>>>
>>> [HEP]
>>>
>>> [Ring_1]
>>> 19 20 21 22 23 24 25 26 27 28 29 30
>>> 32 34 37 38 39 40 42 44
>>>
>>> [Ring_17]
>>> 979 980 981 982 983 984 985 986 987 988 989 990
>>> 992 994 997 998 999 1000 1002 1004
>>>
>>> Those two rings are the the aggregates in the VMD. I indentified them 
>>> and does the g_dist
>>> between the two but my data looks a little bit off.
>>>
>>> g_dist -f PAP_20ns.xtc -s PAP_20ns.tpr -n PAP.ndx -b 19000 -e 20000 
>>> -o result.xvg
>>>
>>> The value I got is big and does not look right. The diatnce is around 
>>> 3.5 nm but if they aggregate it must be a Pi-Pi stacking and should 
>>> be around 0.5 angstrom. Somebody told me i need to develop a script 
>>> to discard some interactions in the pair calculations but did not 
>>> understand what he means ( less ins).
>>>
>>
>> For two non-overlapping groups, this advice does not make sense. 
>> Intramolecular interactions do count for the RDF calculation, though. 
>> Perhaps you've gotten your advice confused?
>>
>> As for the distance, have you corrected for PBC? Are you analyzing the 
>> molecules that you think you are?
>>
>> -Justin
>>
>>> Can you please comment on this please. Your help is really appreciated.
>>>
>>>
>>> Regards,
>>>
>>> Rob
>>>
>>
>> -- 
>> ========================================
>>
>> Justin A. Lemkul
>> Ph.D. Candidate
>> ICTAS Doctoral Scholar
>> MILES-IGERT Trainee
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin
>>
>> ========================================
>> -- 
>> gmx-users mailing list gmx-users at gromacs.org
>> http://lists.gromacs.org/mailman/listinfo/gmx-users
>> Please search the archive at http://www.gromacs.org/search before 
>> posting!
>> Please don't post (un)subscribe requests to the list. Use the www 
>> interface or send it to gmx-users-request at gromacs.org.
>> Can't post? Read http://www.gromacs.org/mailing_lists/users.php
>>
>>
> 

-- 
========================================

Justin A. Lemkul
Ph.D. Candidate
ICTAS Doctoral Scholar
MILES-IGERT Trainee
Department of Biochemistry
Virginia Tech
Blacksburg, VA
jalemkul[at]vt.edu | (540) 231-9080
http://www.bevanlab.biochem.vt.edu/Pages/Personal/justin

========================================