|
Alm, Eric, and Andrew Endy, 20.181 Computation for Biological Engineers, Fall 2006. (Massachusetts Institute of Technology: MIT OpenCourseWare), http://ocw.mit.edu (Accessed 07 Jul, 2010). License: Creative Commons BY-NC-SA
Image of protein Zif28-GCN4, adapted from Wolfe, et al., Crystal Structure of a Zif23-GCN4 Chimera Bound to DNA, in the RCSB Protein Databank. (Figure by Prof. Andrew Endy.)
Course Highlights
This course features many lecture notes and a complete set of assignments with solutions.
Course Description
This course covers the analytical, graphical, and numerical methods supporting the analysis and design of integrated biological systems. Topics include modularity and abstraction in biological systems, mathematical encoding of detailed physical problems, numerical methods for solving the dynamics of continuous and discrete chemical systems, statistics and probability in dynamic systems, applied local and global optimization, simple feedback and control analysis, statistics and probability in pattern recognition.
An official course Web site and Wiki is maintained on OpenWetWare: 20.181 Computation for Biological Engineers.
Technical Requirements
Special software is required to use some of the files in this course: .py, .pdb (Protein Data Bank), and .zip.
Syllabus
Prerequisites
20.180
Assignments and Exams
There will be short assignments distributed at almost every lecture. We expect that all assignments will be turned in by midnight on the day they are due. You may discuss assignments with your classmates but we expect that you will submit your own work. Late assignments (up to one week late) will be given 1/2 credit; solutions for assignments will be posted at the posted due date. A family crisis or severe illness requiring attention from the infirmary and prohibiting you from all your coursework are acceptable reasons for missing an exam; every effort will be made to accommodate you in these exceptional circumstances. More information regarding academic integrity is available here.
Grading
Grading criteria.
|
ACTIVITIES
|
PERCENTAGES
|
|
Problem Sets (Equally Weighted)
|
60%
|
|
Exam 1
|
10%
|
|
Exam 2
|
10%
|
|
Exam 3
|
10%
|
|
Class Participation
|
10%
|
Calendar
The course is divided into three modules, with exams at the end of each module.
Module 1: Phylogenetic Inference (Lec #1-8)
Module 2: Molecular Modeling / Protein Design (Lec #10-17)
Module 3: Discrete Reaction Event Network Modeling (Lec #19-24)
Each module covers the following general topics:
- Data Structure
- Optimization Problem
- Algorithms
Course calendar.
|
LEC #
|
TOPICS
|
KEY DATES
|
|
Module 1: Phylogenetic Inference (PI) (Instructor: Prof. Alm)
|
|
1
|
Course Overview
Introduction to Phylogenetic Inference; Case Studies; Phylogenetic Trees; Quick Review of Recursion
|
|
|
2
|
Review of UPGMA; Purpose of Phylogenetics; Newick Notation
|
|
|
3
|
Phylogenetic Trees: Overview, Possible Trees
Python®: Trees; Data Structure, Parsing Function
|
Homework 1 due
|
|
4
|
Parsimony; Sankoff Downpass Algorithm
|
Homework 2 due one day after Lec #4
|
|
5
|
Downpass (cont.); Fitch's Up Pass
|
Homework 3 due
|
|
6
|
Up Pass (cont.)
|
Homework 4 due
|
|
7
|
Parsimony (cont.); Overall Strategy; Maximum Likelihood (ML); Jukes-Cantor; Evolutionary Model
|
|
|
8
|
Greedy Algorithm for Trying Trees
Review
|
Homework 5 due five days after Lec #8
|
|
9
|
Exam 1
|
|
|
Module 2: Molecular Modeling / Protein Design (MM/PD) (Instructor: Prof. Alm)
|
|
10
|
Introduction to The Protein Design Problem. What Makes Proteins Fold? Entropy
|
Homework 6 due
|
|
11
|
MM/PD Lecture 2
|
|
|
12
|
MM/PD Lecture 3
|
|
|
13
|
Dihedrals, Build Order
|
Homework 7 due two days after Lec #13
|
|
14
|
MM/PD Lecture 5
|
|
|
15
|
MM/PD Lecture 6
|
|
|
16
|
MM/PD Lecture 7
|
Homework 8 due
|
|
17
|
MM/PD Lecture 8
|
|
|
18
|
Exam 2
|
Homework 9 due two days after Exam 2
|
|
Module 3: Discrete Reaction Event Network Modeling (DRENM) (Instructor: Prof. Endy)
|
|
19
|
When to Use Computational Methods vs. Exact Methods; The Physics Model Underlying Exact Methods
|
|
|
20
|
Physics Model Underlying Exact Methods (cont.); Using Physics Model to Compute When a Reaction will Take Place
|
|
|
21
|
Review of How Physics Model Leads to Computational Method; The Complete Computational Method (Gillespie's Direct and First Reaction Methods)
|
|
|
22
|
Difference Between Reaction Rate and Reaction Propensity; Achieving Faster Computation
|
Homework 10 due five days after Lec #22
|
|
23
|
Next Reaction Method Algorithm; Application to Genetic Memory (Latch)
|
|
|
24
|
Review of Key Concepts
|
Homework 11 (optional) due
|
|
25
|
Exam 3
|
|