Lipid as residues topology

From: Jens K. Munk (
Date: Tue Nov 01 2011 - 05:35:37 CDT

Dear NAMD-L,

I would like to make a topology file in which lipids are built like peptides, so that each lipid contains eg. four residues, namely:

1. Head group
2. Phosphoglycerol
3. Acyl chain on glycerol carbon 2
4. Acyl chain on glycerol carbon 3

This would make it much easier to construct any lipid one might be interested in; any combination of acyl chains and head group could be constructed quickly. Also, having defined lipids as made of four residues would make it easier to select eg. chains separately.

I have successfully used this approach to link butanoic acid to phosphoglycerol by defining a BOND between butanoic acid carbon 1 (named C1) and the oxygen on glycerol carbon 2 (named O2), so the BOND definition is


But I need to be able connect two acyl chains and a head group to the phosphoglycerol. The problem is how to define the connection between eg. two butanoic acid molecules and phosphoglycerol, one at glycerol carbon 2 and the other at glycerol carbon 3 *without* having to define two butanoic acid RESI entries, one for each connection.

My guess is it cannot be done.

Is there any way to pass arguments so that I can choose which oxygen to connect to?

I was thinking about defining the bond at glycerol carbon 2 as above and defining the bond at glycerol carbon 3 as a patch. Very big patch though. Any good?

Later I will define the connection between phosphate and the head group by defining

BOND Cx -O12

where Cx is a head group carbon and O12 is one of the phosphate oxygens.

Any thoughts? TIA.

Venlig hilsen / Kind Regards
M. Sc. Jens K. Munk
Department of Basic Sciences and the Environment
Room T432
Phone +45 3533 2429
Thorvaldsensvej 40
DK-1870 Frederiksberg

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:57:53 CST