Dr. Peter Meulbroek

Director, Software Design and Integration, 

Materials and Properties Simulation Center,

Chemistry 139-74, California Institute of Technology

Pasadena, CA 91125

meulbroek@wag.caltech.edu

my pic

Welcome to my home page. I'm a mathematician / economist / geoscientist hiding out in a computational chemistry group, the MSC) at Caltech, currently managing software design and development.  I work for Professor Bill GoddardHere's a pointer to my old Grad Student web page at Cornell, if you care.

Note: recent changes including moving a lot of the code development information to a new server at ruby.wag.caltech.edu. If a link is dead, just go there...

My research interests are:

Applications and Database Design and implementation.  I now manage several software developments (co-manage;  the science aspects are handled by my fellow MSC directors).  These include

VLS:  virtual ligand screening.   Scanning for a ligand that binds with a protein receptor target is not only algorithmically and scientifically challenging, it also presents severe technical design issues.  We have to be able to test and reject a ligand as a potential binder within about 2 minutes of calculation so that we can process our entire 1/2 million member ligand database on a reasonable number of machines.  We have to manage those jobs, manage the scientific code that does the test, and keep everyone informed about what is happening.  The  structure for VLS is shown here.

CMDF:  the computational materials design facility.  A new project here at the MSC that involves designing and implementing a facility that can perform, organize, and utilize chemical simulations on a large number of length scales (from QM to continuum), over symmetric time scales (from pico-seconds to seconds).  

The EOS Interface worksheet!!!  Liquid Vapor equilibria in a dynamic excel environment. 

Use of Equations of State for Hydrocarbon Exploration.  Notes from a recent series of articles in Organic Geochemistry.

The Hydrocarbon Toolkit. Note that there are now two versions for the HC Toolkit: the original, developed in Mathematica, it is a collection of routines to calculate liquid-vapor equilibria for organic mixtures.  These routines are currently in stasis, as I do not do much mathematica programming any more. The sourceforge site for it is here. The second is a set of Perl routines (acutally, translations of the mathematica to perl) that do the same things. These are currently maintained, and quite interesting. Check them out at the ruby EOSInterface page

Quaterions:  from a obscure 19th century invention to THE way to do fast, accurate, interpolateable graphics, these beasties have had quite a journey.  As a side note, I recently spent some time understanding them and their graphical implications.  The results of the study were a Mathematica package update and a help file, bundled for your convenience.  

Database Design and Implementation.  Using databases to store large numbers of data (from Experiment or Simulation), and to mine that data to predict materials properties.  

A smiles parser.  The smiles language is an extraordinarily compact way of expressing molecular structure.  Its one shortcoming is that it doesn't include atom positional information, which makes viewing the structures problematic.  I've developed a routine in perl to parse the structure, guess at atom positions, and output them in z-matrix format.  The results can then be fed into a energy minimization program for an accurate 3-D structure.  Some results of the parser can be found here.

Travel.  I have as a serious hobby (it's been called an obsession, but that's not MY word) of keeping tabs on the oligarchy that we call the travel industry.  The hyperlink goes to my personal travel web.

My Links:  every page has to have a list of links, for whatever reason.  Some highlights of mine:  where to find chemical data, how to easily install windows 2K (the ultimate windows, in that after this, I'm definitely Linux bound...)

This page was Last modified: Tue Nov 18 21:53:01 PST 2003 10/21/02.

copyright ©2003 Peter Meulbroek