VMD-L Mailing List

From: Mgr. Lubos Vrbka (lubos.vrbka_at_uochb.cas.cz)
Date: Fri Jan 20 2006 - 15:19:54 CST

hi john,

thank you for your reply.

> A few notes about this. Make sure you don't have any bizarre atom names
> that are confusing VMD, as the 'measure hbonds' code depends on correct
> identification of the atom names currently. A quick test for this is to
> display just the hydrogens with the selection "hydrogen" and make sure you're
> getting what you expect.
well, the names are not bizarre, i think - this is special water model
where oxygen is OW, hydrogens are HW1 and HW2, and extra points are EP,
LP1, LP2 (this is 6-center water model). iirc, both HW1 and HW2 are
reported as 'hydrogen'. i can't check it right now unfortunately...

> The order of $selection1 and $selection2 matters. If you provide both,
> then $selection2 is treated as the list of acceptor candidates.
right, this agrees with the documentation.

> When using 'measure hbonds', you should avoid having the same atom listed
> in both the $selection1 and $selection2 parameters if you desire to have
> reliable results, there's a warning about this in the manual as you've
> already noted.
in my case
"name OW" gives the list of all oxygen atoms
"index 5" is one of them
so this violates the rule mentioned above :( however, i don't know how
to ensure that the one oxygen of interest isn't contained. my intention
would be probably (pseudo-code)
foreach x [atomselect "name OW"]
measure hbonds 3.0 30 x ([atomselect "name OW"] - x)
measure hbonds 3.0 30 ([atomselect "name OW"] - x) x
how can it remove from the list? then i might avoid the problem.

i wanted to download the files to my computer and download vmd (i'm at
home now and don't have vmd installed here), however i'm not able to
download vmd at the moment... maybe later...

if the issue i am observing would be reproducible and i could rely on
it, it would make my life easier, since i wouldn't have to do 2 measure
hbonds operations (one for the x as a donor and the other for the x as
an acceptor):

>> measure hbonds 2.9 30.0 [ atomselect top "index 5" ] $ow
>> 5 is one of the oxygens contained also in the 'ow' list (all oxygens)
>> this gives me correct answer (i checked it in the structure)
>>
>> {5 5} {191 473} {1 0}
in this case it reports 2 donor hbonds, iirc

>> however when i reverse the order
>> measure hbonds 2.9 30.0 $ow [ atomselect top "index 5" ]
>> i get "more than i expected":
>>
>> {989 5 5 815} {5 191 473 5} {989 1 0 810}
this gives 2 donor and 2 acceptor hbonds. it's true, this is ice lattice.

do you think i can rely on this, or should i do some trick (the question
above) to always remove the respective oxygen from the ow list?

thank you.with best regards,
lubos 

-- 
.....................................................
Mgr. Lubos Vrbka
Center for Biomolecules and Complex Molecular Systems
Institute of Organic Chemistry and Biochemistry
Academy of Sciences of the Czech Republic
Prague, Czech Republic
http://www.molecular.cz/~vrbka
.....................................................