Topic Options
#15463 - 09/07/07 12:29 PM inter intra hbond
beginner Offline
Forum Member

Registered: 07/19/04
Posts: 165
Dear all,

I would like to calculate the average number of inter and intra- hydrogen bond for my lipid bilayer. How can I compute them ?

Thank you very much.

Top
#15464 - 09/07/07 01:04 PM Re: inter intra hbond [Re: beginner]
rmv Online   content

Forum Member

Registered: 09/17/03
Posts: 8373
Loc: 39 03 48 N, 77 06 54 W
COOR HBOND (corman.doc) with the appropriate atom selections ...

Top
#15465 - 09/07/07 01:24 PM Re: inter intra hbond [Re: rmv]
beginner Offline
Forum Member

Registered: 07/19/04
Posts: 165
Yes, but all the lipid residues have the same atom type.
I was thinking about to specify the atom type and residue number.
However, how can I make it fast ?

Thank you very much.

Top
#15466 - 09/07/07 02:20 PM Re: inter intra hbond [Re: beginner]
rmv Online   content

Forum Member

Registered: 09/17/03
Posts: 8373
Loc: 39 03 48 N, 77 06 54 W
The appropriate atom selections depend on the lipid topology. For most lipids, intra-lipid H-bonds are rare. For PC lipids, it may be more correct to think of choline:phosphate ion pairs, rather than H-bonds. Lipids are good H-bond acceptors, esp. the PO4 group. Note that only more recent CHARMM versions have image support for COOR HBOND, which is needed for lipids near the unit cell boundaries.

Top
#15467 - 09/07/07 03:48 PM Re: inter intra hbond [Re: rmv]
beginner Offline
Forum Member

Registered: 07/19/04
Posts: 165
Sorry, I should make it more clear.
I would like to compute the intra hydrogen bond of PE head group. if I say "coor hbond sele type P end sele type hn* end", I will get both inter and intra hydrogen bond. I wonder if it is possible to make it faster by adding a special key word.

As you mentioned
Quote:

Note that only more recent CHARMM versions have image support for COOR HBOND, which is needed for lipids near the unit cell boundaries.


, could you please tell me more how can I include the image into the calculation.
Thank you very much.

Top
#15468 - 09/07/07 05:58 PM Re: inter intra hbond [Re: beginner]
rmv Online   content

Forum Member

Registered: 09/17/03
Posts: 8373
Loc: 39 03 48 N, 77 06 54 W
Assuming DLPE, and SEGID L of 80 lipids, something like

DONO ADD SELE ATOM L * HN+ END
ACCE ADD SELE ATOM L * O1+ END
SET K 1
LABEL RESLP
! inter lipid
COOR HBOND SELE ATOM L @K HN+ END -
SELE ( TYPE O13 .OR. TYPE O14 ) .AND. .NOT. ATOM L @K * END
! intra lipid
COOR HBOND SELE ATOM L @K HN+ END -
SELE ( TYPE O13 .OR. TYPE O14 ) .AND. ATOM L @K * END
INCR K BY 1
IF K LE 80 GOTO RESLP

It probably won't be particularly fast. Since most lipids don't have H-bond DONOr and ACCEptor statements in the RTF, they may need to be explicitly added, as above.

For image support, c33 or later is required; for details, see the COOR HBOND description near the end of corman.doc

Top

Moderator:  John Legato, lennart