[gmx-users] analysis of POPC

[chris.neale at utoronto.ca]

> There are fewer data for POPC than for DOPC and DPPC.  If your force  
> fields and methods give you good results for DOPC and DPPC, you  
> should be in good shape for POPC, since the building blocks are the  
> same, but just rearranged.

Agreed. I add my emphasis to the given reference to the DOPC/DPPC  
combination to ensure that your double bond is correct (A quick look  
at the topology from the tieleman website near the double bond  
indicates that this is not at all evident and should be tested as  
suggested by the previous post.)

A simple google search for "Nagle popc" (without quotes) indicates  
that the experimental information you are looking for is easily  
available. I have used popc in the past and found it to basically meet  
the experimental data (based exactly on the topology available in the  
tieleman website, and ignoring my inability to understand exactly why  
the double bond was treated as it was). That said, it is my current  
opinion that the Charmm parameters are superior for lipid in the  
absence of protein. I don't use the charmm parameters myself, but from  
the papers I read the charmm lipid parameters appear to be better (my  
understanding, and definitely up for debate)

Chris.