[gmx-users] g_hbond and contact
Erik Marklund
erikm at xray.bmc.uu.se
Tue Mar 5 09:56:28 CET 2013
To clarify: the -hbn output is not very indicative of how many
contacts there were since some of them could be present in one frame
but absent in 999999. The -num option, however, provides the number of
contacts over time, and its time average probably tells you much more
in this case.
What version of g_hbond are you using? I remember there were several
bugfixes over the last 6 months or so. With the latest version(s) I
believe that the -merge flag has no effect on contact analysis, which
is correct.
Erik
On Mar 5, 2013, at 5:20 AM, Kavyashree M wrote:
> I am sorry There is no "7ws" Its a typographic error. What I wanted
> to ask
> was -
> I am comparing two simulations S1 and S2 for contacts at a given cut
> off
> I used g_mdmat and g_hbond to calculate it.
> g_hbond outputs: of "-num" and "-hbn" was considered.
> 1. -hbn output was analysed to calculate how many
> contacts each
> atom has from both S1 and S2.
> 2. -num output graph was compared from both S1 and S2.
> g_mdmat output: of "-no" was considered.
> 3. -no output was analysed from both S1 and S2 using the
> third column or the second Y value which gives total
> contacts
> of each atom.
>
> It was observed that "1" and "3" matched exactly giving the same
> number of
> contacts each atom has (in the whole simulations). indicating that the
> number
> of contacts each atom has was more in S2 than S1.
>
> But the graph from "2" indicated that the number of contacts (along
> the
> trajectory)
> in S1 was higher than S2.
>
> My doubt is:
> The number of contact per atom follows S2 > S1
> while number of contacts per time follows S1 > S2.
> I am unclear as to what I have to conclude from this observations.
>
> -----> I used the same cutoff throughout.
> -----> There has not been any swapping of the trajectory while
> analysing.
>
> Thank you
> Kavya
>
>
> On Tue, Mar 5, 2013 at 1:09 AM, Justin Lemkul <jalemkul at vt.edu> wrote:
>
>>
>>
>> On 3/4/13 1:10 PM, Kavyashree M wrote:
>>
>>> On Mon, Mar 4, 2013 at 11:10 PM, Justin Lemkul <jalemkul at vt.edu>
>>> wrote:
>>>
>>>
>>>> When measuring contacts, you don't measure one group, you measure
>>>> the
>>>> number of contacts that occur between groups A and B, which
>>>> considers all
>>>> atoms in those two groups.
>>>>
>>>>
>>> I gave a group of hydrophobic atoms in both cases
>>> The command I gave -
>>> g_hbond_46 -f x.xtc -s x.tpr -contact -n x.ndx -
>>> r 0.4
>>> -hbm o.xpm -hbn o.ndx -num o.xvg
>>> my index file contained a group of hydrophobic atoms. which I
>>> supplied in
>>> the x.ndx.
>>>
>>>
>>>
>>>> You don't define contacts in an index group, you define atoms
>>>> that may
>>>> or
>>>> may not make contacts with others.
>>>>
>>>>
>>>> The one I mentioned here is the output index file from the
>>>> g_hbond (4.6
>>> version) - o.ndx.
>>>
>>>
>>>
>>>> [ contacts_C_CA_CB_CD_CD1_CD2_****CE_CE1_CE2_CE3_CG_CG1_CG2_CZ_***
>>>> *CZ2_CZ3_CH2
>>>>
>>>> ]
>>>>> 5 7ws
>>>>> 5 10
>>>>> 5 14
>>>>> 5 18
>>>>> 5 22
>>>>> 5 24
>>>>> 5 27
>>>>> 5 30
>>>>> 5 35
>>>>> 5 292
>>>>> 5 296
>>>>> 5 30
>>>>>
>>>>>
>>>>> There's something very wrong with this index file. How did you
>>>> generate
>>>> it? The presence of a repeated atom number (5) and a nonsensical
>>>> one
>>>> (7ws)
>>>> leads me to believe that you've done something incorrect. Did
>>>> this come
>>>> from g_hbond? It looks like the output of -hbn, which is only
>>>> useful for
>>>> decoding hbmap.xpm, nothing else.
>>>>
>>>>
>>>> I did not generate this. The tool (g_hbond) generated this index
>>>> file.
>>> It
>>> is the -hbn output.
>>>
>>>
>>>
>> OK, then I still don't know what "7ws" is, but the only purpose for
>> this
>> file is to provide a key to the existence matrix in hbmap.xpm. Your
>> previous description indicated that you were using it for some other
>> analysis, which would not be appropriate.
>>
>> The other thing worth mentioning here is something that was posted
>> to the
>> list just a few hours ago, that the output of g_hbond -contact may
>> not
>> agree with other methods of calculating contacts, especially in the
>> case of
>> -merge vs. -nomerge.
>>
>> -Justin
>>
>>
>>
>>>> From this second section Total contacts was extracted for each
>>>> atom and
>>>>
>>>>>
>>>>>> compared with
>>>>> that from a second simulation.
>>>>> These contacts was matching with the contacts of the 3rd column
>>>>> from
>>>>> g_mdmat output -
>>>>> @ legend string 0 "Total/mean"
>>>>> @ legend string 1 "Total"
>>>>> @ legend string 2 "Mean"
>>>>> @ legend string 3 "# atoms"
>>>>> @ legend string 4 "Mean/# atoms"
>>>>> #res ratio tot mean natm mean/atm
>>>>> 1 1.001 11 10.991 1 10.991
>>>>> 2 1.244 10 8.041 1 8.041
>>>>> 3 1.166 13 11.147 1 11.147
>>>>> 4 1.036 11 10.615 1 10.615
>>>>>
>>>>>
>>>>> While the time dependent contacts in the xvg file shows that the
>>>>> first
>>>>> simulation has more
>>>>> contacts than the second one..
>>>>>
>>>>>
>>>>> That shouldn't be unexpected. Two independent simulations have no
>>>> guarantee of doing the same thing, that's why sampling is so
>>>> important.
>>>>
>>>>
>>> Thank you
>>> kavya
>>>
>>>
>> --
>> ==============================**==========
>>
>> Justin A. Lemkul, Ph.D.
>> Research Scientist
>> Department of Biochemistry
>> Virginia Tech
>> Blacksburg, VA
>> jalemkul[at]vt.edu | (540) 231-9080
>> http://www.bevanlab.biochem.**vt.edu/Pages/Personal/justin<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<http://lists.gromacs.org/mailman/listinfo/gmx-users
>> >
>> * Please search the archive at http://www.gromacs.org/**
>> Support/Mailing_Lists/Search<http://www.gromacs.org/Support/Mailing_Lists/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/**Support/Mailing_Lists<http://www.gromacs.org/Support/Mailing_Lists
>> >
>>
> --
> 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/Support/Mailing_Lists/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/Support/Mailing_Lists
More information about the gromacs.org_gmx-users
mailing list