Molecular Modeling

and Simulation

09-446/09-862

(These are fictional numbers.

Of course no one would let a

graduate student teach a

class at CMU)

Syllabus: Fall 1998

Units: Graduate credit 12

Undergraduate 9

1. Hours:

1.1 Lecture: TR 10:30 – 11:20, MI553

1.2 Lab: MW 18:00 – 19:30, SGI cluster 3^rd floor of MI

1.3 Office Hours: TR 12:30 – 14:00

or by appointment

held @ 509A Mellon Institute (MI)

2.Contact information:

2.1 Ivaylo Ivanov

     P.O. Box #49

     Mellon Institute

     4400 Fifth Ave.

     Pittsburgh, PA 15213

2.2     e-mail: iivanov@andrew.cmu.edu

2.3  www.andrew.cmu.edu/user/iivanov

weber.chem.cmu.edu

2.4   tel : (412)268 1054

office: 509A

2.5 Remarks:

* the preferred way to contact me is by e-mail

* or you can just drop by my office @ Mellon (if you do this in the suggested Office hours you are guaranteed to find me there)

*please, do not use the telephone number except for extreme emergencies!

3. Computer accounts on the SGI cluster will be set up for each student. Disk space is limited to 30Mb /account. In addition a limited amount of supercomputing time will be available for the group projects only. Students intending to make use of this should contact me at a later date (but certainly prior to the project submission deadline)

4. There is no required textbook for this course. Students are certainly encouraged to read from the optional textbook:

Andrew Leach , Molecular Modeling: Principles and Applications.

The book has a recent and very extensive bibliography on many of the topics covered in this course and may serve as a good starting point in researching for the group projects. It is available at the bookstore.($45.00)

Other books that may be of interest, especially for those taking the course for graduate credit:

D. Rappport , The Art of Molecular Dynamics Simulation

Allen & Tildesley , Computer Simulation of Liquids( On course reserve in the MI Library )

In addition a number of research papers will be used in the course. The papers and other relevant materials will be available from the course web site.

Students less familiar with Unix may need to consult a standard book , for instance :

M. Sobell Unix System V: A Practical Guide

The online Polaris help system could also be useful.

5.0 Purpose and Expectations for the course.

This course presents an overview of the contemporary molecular modeling and simulation methods.It is intended primarily for chemistry, physics, the life sciences, and material science majors but other students are more than welcome to attend. The lectures will address the physics underlying the different simulation techniques , while the lab should provide the necessary practical experience in dealing with calculations and introduce some popular simulation packages. The design of the course is intended to provide the proper balance between theory and applications and to this end several interesting problems ranging from NMR refinement of enzymes to the study of Langmuir Blodgett films to DNA base pair interactions and DNA chips will be considered . Ideally, at the end of the course the students will be able to critically evaluate the different approaches and algorithms in terms of their applicability to specific chemical problems, assess the limitations of the methods, read and evaluate theoretical papers, perform calculations using packaged code and interpret the results. The final projects would emphasize collaborative work on a topic chosen to reflect the students’ individual interests and development of presentation skills.

6. Prerequisites

The course requires some level of mathematical sophistication ,including but not limited to:

A. Calculus + Differential Equations

B. Elementary Probability and Statistics

Some important mathematical concepts that not ordinarily included in the typical undergraduate curriculum will be introduced during lectures.

If you are in doubt check out the quiz available on the course web site. If you can answer 60% or more of the problems correctly there is a high probability that you will do well in the course.

As far as physics is concerned minimal familiarity with concepts from Quantum Mechanics will be expected. Therefore it would be preferable but not required that you have taken :

C. Introductory Quantum Mechanics as taught

in undergraduate chemistry or physics courses.

For graduate students only:

D. Programming in a structured programming language

like FORTRAN, C or C++ ( equivalent to 15-127)

7. Schedule

Lecture 1 Mathematics of Computational Modeling

Some mathematical prerequisites from the Calculus of Variations

Minimization of Energy Functions

Numerical Exploration of Energy Functions

Eigenvalue Problems in Hilbert Space

Phase Space & Statistical Mechanics

Lecture 2 Review of Thermodynamics and Statistical Mechanics

Lab: Week 1 – No Lab

Lecture 3 Molecular models. Molecular Mechanics. Empirical Force Fields. Energy Minimization .

Lab 1: Host Guest Chemistry. Energy Minimization calculations on crown ether with different alkali metal ions.

Lectures 4, 5, 6 Molecular Dynamics:

Introduction

Integration Algorithms (Verlet ,Velocity Verlet , Leapfrog, Predictor-Corrector Methods )

Periodic Boundary Conditions

NVE ,NVT ,NVP calculations. Thermostats and Barostats.

Long range interactions. Efficient Algorithms for Electrostatics

Lecture 7 Conformational Analysis

Lab 2: Molecular Dynamics refinement of the NMR structure of pancreatic ribonuclease A with solvent molecules.

Lecture 8 Configurational Sampling

Lecture 9, 10 Monte Carlo. Path Integral Methods.

Case Study: Configurational Bias Monte Carlo.

Lecture 11,12 Brownian Dynamics. Langeven and Focker –Plank Equations. Correlation Functions.

Lecture 13 Free Energy Calculations (Computational Alchemy)

Lab 3: Simulation of Langmuir Blodgett films with Molecular Dynamics and configurational

bias Monte Carlo.

Lecture 14 Direct and Indirect Drug Design Methods. (QSAR, docking, de novo methods, 3D database searching , etc)

Lectures 15, 16, 17 Quantum Mechanics

Ab initio Methods

Semiempirical Methods

Density Functional Theory

Lab 4: Reaction Dynamics. SN2 reactions.

Lecture 18, 19 Hybrid Quantum/Molecular Dynamics

Case Study: The Car-Parrinello Approach to Molecular Dynamics

Lab 5: INDO and PPP calculations on polyacetylene . Applications .

Lectures 20, 21 Critical Evaluation of Theoretical Papers

Lab 6: Study of the splitting of levels in oligonucleotides by density functional theory.

Lectures 22-26 These would allow for some flexibility in covering the topics. One would be held as a review before the midterm exam.

The Lab schedule above is for undergraduate students only and will be posted on the Web site as well. In general the lab assignments will concentrate on the use of packaged code.

Four additional programming assignments in a high level structured programming language (FORTRAN,C,C++) will be handed to students taking the course for graduate credit. These are as follows:

Implementation of a simplified version of the FMM and Barnes Hutt algorithms. Parallelizing the code (optional for extra credit).

Particle Mesh Ewald method for electrostatic interactions.

Calculation of the dynamics of a Cl 2 molecule at the surface of a Si monocrystal by Car Parrinello Molecular Dynamics.

Computational Alchemy. Umbrella sampling.

9.0 General requirements

You may skip a lecture or two but you will be expected to know the material.

You are expected to attend all Labs.

Every student will have to complete at least 6 homework assignments. Collaboration on these should be minimal (limited to discussion of the homework). Homework is due ONE WEEK after you get the assignment. Further extension for one week is possible but you get only half the points.

All of you are required to complete an individual or group project on a topic relevant to the course and reflecting your interests in the field. Project topics are subject to approval. During the week following the midterm exam I will ask you to submit a one-page abstract or meet with me to discuss the project. Your grade for the project will reflect both the accomplished work and the quality of the presentation. For group projects each member will be evaluated on his/her individual contribution which should be readily identifiable.

10.0 Additional Requirements for Graduate Students

Four additional programming assignments.

Definition of cheating:

1 Any kind of collaboration during the Midterm exam.

2 Sharing of code for grad students (discussion is always OK)

3 Copying of homework assignments or using somebody else’s results to write the assignment

Enjoy!