AMBER Archive (2006)

Subject: Re: AMBER: mixing GAFF and CHARMM

From: Ravinder Abrol (
Date: Wed Dec 20 2006 - 11:38:23 CST

Thanks Ran. I was refering to GAFF for ligand, as I have CHARMM parameters for lipid.
But your arguement still applies.
If someone has used GAFF for ligand and CHARMM for protein, even as a test case,
I would be interested to hear about how well it worked.

----- Original Message ----
From: <>
Sent: Wednesday, December 20, 2006 9:12:05 AM
Subject: Re:AMBER: mixing GAFF and CHARMM

Dear Ravi
I hope this can give you some kind of help:
"The statement that one should use GAFF for protein, water, if GAFF is used for
lipid, may not be correct. Lipids and proteins can have different parameter sets,
just as you can use "standard" AMBER parameters for proteins and use your favorite
parameters for lipids (you can even develop your own if you so inclined). In
particular, GAFF referes to a set of parameters for modeling of general organic
compounds. This set has not been tuned specifically for proteins (or peptides), as
such there is not much detailed studies on the application of GAFF to model
proteins. So, you are better off to use "standard" protein parameters for proteins
and use GAFF for others. Again, there is no inherent conflict between these sets.
It is a matter of accuracy. If you are interested in protein dynamics in membrane,
you should use the protein force fields for proteins. As for lipids, you can use
others, including GAFF. Regardless which specific set you choose for the lipid,
please make sure the lipids confirm to experimental observations."
---you can find this statement at AMBER Archive
I'm also simulating a system containing lipid bilayer and don't yet have a good
idea about using amber to solve this problem.
Ran Cao

The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to

Do You Yahoo!?
Tired of spam? Yahoo! Mail has the bes
The AMBER Mail Reflector
To post, send mail to
To unsubscribe, send "unsubscribe amber" to