[discuss ifpsc] Re: [ifpsc] Re: Forcefield database proposal resulting from our

Jonathan: Thanks a lot for bringing up the discussion to this mailing list!

I agree that both type (general and specific) force fields should be included in the database. I also think Ray's suggestion is excellent. We can use Blog methods, for example, to allow people publish his/her force field parameters, others can put comments or even revisions, but cannot change the previous data. To create and maintain such a web-based database does not cost much. Anyone can add or download the data freely (maybe need a little police work?). But using these force fields requires some knowledge on how to make the parameters actually work with a simulation engine.

For casual or non-professional modelers, however, to figure out how to make the parameters work is not a trivial task, much worse time is when some parameters are missing. Those users want "push a button" operation. The force field must be reliable and accurate. Those general and published force fields such as AMBER, OPLS, CFF, COMPASS can be in this category. The database should assign atom types and force field parameters correctly for a given molecule, for a selected simulation engine, automatically.

If the coverage is not enough, which is often true, the database should be extensible with the same quality. This is the central part of my proposal. To make and maintain such a database, however, needs financial supports.

I think, any one of these two methods will probably not solve all problems, but both may.

Huai


----- Original Message ----- From: "Moore, Jonathan (J)" <JMoore2@xxxxxxx>
To: <discuss-ifpsc@xxxxxxxxxxxxx>
Cc: "Huai Sun" <huaisun@xxxxxxxxxxx>
Sent: Thursday, September 13, 2007 5:48 AM
Subject: [ifpsc] Re: Forcefield database proposal resulting from our


Huai,

Thanks for your response.  My preference would be to discuss the
forcefield database ideas on the discuss-ifpsc@xxxxxxxxxxxxx public
email list (it's nominally public, but right now it is the IFPSC
steering team...though I took the liberty of adding you).  That way,
other IFPSC folks who are interested can be involved in the discussion.

I take your point that existing force fields are too diverse to be used
together.  My vision isn't of a database with all the parameters (or
even most of them) are suitable for use with each other.  I imagine
something more heterogeneous including the major force fields (AMBER,
Gromos, TraPPE, COMPASS, TEAM, etc.) there as well as many much more
limited models (maybe only applicable to one molecule).  Though general
and transferable is the ideal, both types (general/transferable and not)
are part of the literature and both have some usefulness.  Admittedly,
I'm sure I don't fully appreciate how difficult it would be to integrate
such a diverse set of force fields into a database that would be useful.

Jonathan

____________________________
Jonathan Moore, Ph.D.
Research and Engineering Sciences - New Products
Core R&D
The Dow Chemical Company
1702 Building, Office 300E
Midland, MI 48674  USA
Phone:  (989) 636-9765
Fax: (989) 636-4019
E Mail: jmoore2@xxxxxxx

-----Original Message-----
From: Huai Sun [mailto:huaisun@xxxxxxxxxxx]
Sent: Sunday, September 09, 2007 1:10 AM
To: Moore, Jonathan (J)
Subject: Re: [ifpsc] Forcefield database proposal resulting from our

Dear Jonathan,

First of all, thanks a lot for your kind email and suggestions. I just
came
back from vacation, that's why I am slow in response. We are studying
all
the comments and suggestion received and revising our proposal.

Generally speaking, I appreciate the idea of open, user-supported
databases.
My concern is that existing force fields are too diversified to be put
together. Basically, the atom types are named differently, the
parameters
for the same types of interactions (e.g. C-C-C angle terms) are
differrent
from different developers. The consequence is that these parameters can
be
used only for the systems (molecules) they are designed for; for most
industrial uses, the coverage is not enough.

Based on my experiences, in order to make a high quality and
transferrable
force field, the parameters must be internally consistent. The same
difinition of atom types and same protocol of parameterization must be
used.

We can create a force field repository that users can put and get
freely. To
create such a website and database does not cost too much, we can do it
on
our own expenses. Meanwhile, we are developing consitent and extensible
databases from literature and from our own work. For that we have and
will
continue put in a lot of efforts. However, I believe only the letter can

solve the industrial needs. We are working on it currently, by the time
of
next IFPSC meeting in Novermber, I hope to show a protocol. For
continuring
developments and supports we must find someway to support this effort.

At this point, in order to find the best way to go, I am discussing the
ideas with folks I know.

Best Regards,

Huai


----- Original Message ----- From: "Moore, Jonathan (J)" <JMoore2@xxxxxxx>
To: <ifpsc@xxxxxxxxxxxxx>
Cc: <huaisun@xxxxxxxxxxx>
Sent: Wednesday, August 08, 2007 3:36 AM
Subject: RE: [ifpsc] Forcefield database proposal resulting from our


Fiona,

I think Huai's proposal is, in general, excellent.  If it is realized,
I
think it will be a great help.  I hope he will be successful.

However, my enthusiasm is significantly dampened by its fee-based
nature.  I would be much more interested in a plan that required
monetary support to get it up and running (building the database, the
website, etc.) but then transitioned to a free and open resource where
the users (rather than Huai's group alone) were responsible for
uploading parameters, validation studies, etc.  The forcefield
developers themselves would take on the small burden of uploading
their
parameters, etc. via a user-friendly interface provided by the system.
When parameters aren't available or aren't adequate, the user could be
directed to options for consulting or software purchase, but I'm more
interested in a database system that is open and maintained by the
community than one that is proprietary and only accessible after
payment.

That direction (in the spirit of Wikipedia, Free/Libre/Open-Source
Software, open access in science, etc.) is what I'd like to promote.

Jonathan


Jonathan

____________________________
Jonathan Moore, Ph.D.
Research and Engineering Sciences - New Products
Core R&D
The Dow Chemical Company
1702 Building, Office 300E
Midland, MI 48674  USA
Phone:  (989) 636-9765
Fax: (989) 636-4019
E Mail: jmoore2@xxxxxxx

-----Original Message-----
From: ifpsc-bounce@xxxxxxxxxxxxx [mailto:ifpsc-bounce@xxxxxxxxxxxxx]
On
Behalf Of Fiona Case
Sent: Tuesday, August 07, 2007 12:55 PM
To: ifpsc@xxxxxxxxxxxxx
Cc: huaisun@xxxxxxxxxxx
Subject: [ifpsc] Forcefield database proposal resulting from our

Hello,

Here is the proposal from Huai Sun that I mentioned at our committee
meeting. I think this is a great idea. I have at various times in my
career
spent many days/weeks searching for force field parameters and typing
them
into files - and I occasionally had to abandon projects because good
parameters were not available. If webDDF had been available I would
have

certainly used it!

Huai would be very interested to receive feedback from the IFPSC
committee.
His email is

"Huai Sun" <huaisun@xxxxxxxxxxx>

Best wishes

Fiona

_________________________________________________________________
Learn.Laugh.Share. Reallivemoms is right place!
http://www.reallivemoms.com?ocid=TXT_TAGHM&loc=us
========================================================

The Industrial Fluid Properties Simulation Collective is an international collaboration between industry, academia, and national labs aimed facilitating and promoting the use of molecular simulation methods for the prediction of fluid properties.

discuss-ifpsc@xxxxxxxxxxxxx is an email list for discussing issues related to the Industrial Fluid Properties Simulation Collective...fluidproperties.org

If you are not subscribed, you can do so by sending email to discuss-ifpsc-request@xxxxxxxxxxxxx with 'subscribe' in the Subject field

You can unsubscribe by sending email to discuss-ifpsc-request@xxxxxxxxxxxxx with 'unsubscribe' in the Subject.

Online, searchable archives of this list are available at http://www.freelists.org/archives/discuss-ifpsc


========================================================

The Industrial Fluid Properties Simulation Collective is an international 
collaboration between industry, academia, and national labs aimed facilitating 
and promoting the use of molecular simulation methods for the prediction of 
fluid properties.

discuss-ifpsc@xxxxxxxxxxxxx is an email list for discussing issues related to 
the Industrial Fluid Properties Simulation Collective...fluidproperties.org

If you are not subscribed, you can do so by sending email to  
discuss-ifpsc-request@xxxxxxxxxxxxx with 'subscribe' in the Subject field

You can unsubscribe by sending email to discuss-ifpsc-request@xxxxxxxxxxxxx 
with 'unsubscribe' in the Subject.

Online, searchable archives of this list are available at 
http://www.freelists.org/archives/discuss-ifpsc

Other related posts: