Physics 434, 2014: Syllabus

Back to the main Teaching page.

Back to Physics 434, 2014: Information Processing in Biology.

Course Title

Physical Biology: Information Processing in Biological Systems

Course Numbers

Physics 434/Biology 434/NBB 470/Physics 731, Fall 2014, Emory College

Professor

Ilya Nemenman ilya.nemenman@emory.edu (best way to contact me)

Office Hours

Thu 4-5pm and by appointment

Class Web Page

http://www.menem.com/~ilya/wiki/index.php/Physics_434,_2014:_Information_Processing_in_Biology

Textbooks

There is no required single textbook that will cover all subjects discussed in the class. The book with which the class has the strongest overlap is the following

1. P Nelson, Physical Modeling of Living Systems, WH Freeman, Dec 2014

While the book is not yet published, I have an advanced copy of the book, and will be distributing relevant sections of the book in paper format as needed. We will additionally rely on the following books, and I suggest you borrow them from the library or from me, and read before bedtime.

1. Introduction to Probability, by CM Grinstead and JL Snell. Available at http://www.dartmouth.edu/~chance/teaching_aids/books_articles/probability_book/pdf.html
2. Physical Biology of the Cell by R Phillips, J Kondev, J Theriot (Garland Science, 2008)
3. Biological Physics: Energy, Information, Life by P Nelson (W.H. Freeman, 2003)
4. Biophysics: Searching for Principles, by W Bialek, available at http://www.princeton.edu/~wbialek/PHY562.html

Additional recommended reading includes:

1. Theoretical Neuroscience: Computational and Mathematical Modeling of Neural Systems by P Dayan and L Abbott (MIT Press, 2005)
2. Random Walks in Biology, by H Berg (Princeton UP, 1993)
3. An Introduction to Systems Biology: Design Principles of Biological Circuits by U Alon (Chapman and Hall, 2006)
4. Spikes: Exploring the Neural Code by F Rieke, Dd Warland, R de Ruyter van Steveninck, W Bialek (MIT Press, 1999)
5. E. coli in motion by H Berg (Springer, 2003)

Required Software

Mathworks Matlab Student Edition (free on Emory computers) or any version or Octave v 3.4 or greater (Open Source) is required.

Pre-requisites

• Calculus (Math 111/112 or Math 115/116 or AP equivalent)
• Intro Physics (Phys 141/142 or Phys 151/152 or AP equivalent)
• Intro CS (CS110 or CS170 or instructor consent)

Recommended prior classes:

• Intro Biology (Biol 141/142)
• It is recommended, but not required, that the students have some exposure to differential equations, probability, and statistics.

This class is an upper division physics class, and it will involve advanced mathematical and physical concepts comparable to other upper division physics classes. However, no specific prerequisites beyond those indicated above are required, and the necessary concepts will be introduced within class as needed.

This class differs from most other classes in that we will use mathematical modeling and computer simulations as a language that glues physicists and biologists within the class. Computer simulations will help us to achieve deep, physics-style understanding of certain biological phenomena. Therefore, it is important that you come ready with the needed computational background, in addition to math. I don’t expect you to be professional programmers, but it will be very useful if you can write a simple computer program in your favorite computer language that would output a “Hello world!” sentence on a screen. There will be study sessions for the class where you will be able to remind yourself how to program, and where we will review the needed math with you.

I expect that, working in groups, those of you with biology backgrounds will learn ideas of computing / modeling from your physics / mathCS peers, and the physicist will learn basic biology facts from the biologists.

Class structure

I request a lot from my students, but also provide them with all the necessary resources to succeed in the class. As a result, students will work hard, but will likely learn more than in a typical class. In practice, this philosophy is implemented as follows.

Lectures

The class will be delivered in a traditional lecture form. However, each lecture will start with a set of questions that will test your understanding of the previous material. We will answer the questions collectively, and I expect that everyone will participate in the ensuing discussion. The questions will not be graded – so don’t be afraid to answer incorrectly. Participation is key here.

Homeworks

We will have weekly or biweekly homework problems. The assignments will be revealed typically online on a Tuesday or on a Wednesday. They will be due on a Wednesday (note the change) of the following week. The assignments will involve calculations and problems, like in physics and math classes, and numerical simulations, like in computational physics, biology, and chemistry classes. I expect that, to earn a high grade on your homework assignments, most of you will need to commit about 6 hours a week to them if you choose to work in groups and to attend support sessions. You may need to work a lot more if you choose to do it alone. If you start working on an assignment the night before it’s due, you won’t have time to finish. Don’t be discouraged if you cannot figure out a solution to a homework problem immediately: talk to your peers, attend review sessions, or see me. The problems are meant to be challenging, but, as prior years have shown, they are all solvable if you expend a sufficient effort.

Some of the problems will be open-ended, and may result in research projects if studied deeply (see below). The open-ended sections will be marked clearly and will be graded for extra credit only. The course will be structured to accommodate both undergraduate and graduate students by providing two sets of homeworks with the appropriate levels of difficulty.

Review Sessions and office hours

I will have a regular office hour every week. In addition, we will have review sessions at a time TBD. The sessions will not be required, but attending them is one of the most efficient ways to learn. The time for the sessions will be chosen after a class survey.

Exams

We will have an in-class midterm exam, with a structure similar to other upper division physics classes. There will be no final exam per se; instead every student will need to complete and present a project and write a report (see below).

Projects

Midway through the class, I will reveal a set of open-ended projects. Working in groups of 3-4 people, you will choose one of such projects, and you will work on it for the rest of the semester. These will be open-ended problems, with multiple steps of increasing difficulty, progressing from simple review of class material, to research questions (though those parts of the projects where I don't know if a solution exists will, clearly, not be graded). During one of the last class meetings, the groups will present the project work to the rest of the class. Each student will then be responsible for an individual written report on the project, to be submitted during the exam week. For this report, I will assign you additional questions/problems related to it, and this part of the report will serve as a take-home exam of sorts. I hope that some of these projects may result, in due time, in research papers -- but this has only happened once in many years of teaching this class. On the other hand, an optimist would say that this is known to have happened!

Rescheduled classes

I will need to travel professionally during the semester. As a result, a few classes will need to be rescheduled, or conducted by guest lecturers. These are

1. September 18, to be rescheduled
2. October 28, 30 -- may still happen as scheduled. If not, they will be rescheduled.
3. November 13 -- may still happen as scheduled. If not, a guest lecturer will be invited.
4. December 9 -- a guest lecturer will be invited.

Grading

• Homework assignments – 50%
• Midterm exam -- 10%
• Project work and project presentation -- 25%
• Project writeup / take-home questions -- 15%

In-class questions will not be graded. There may be extra-credit problems available on occasion, which you can use to boost your grade. Your scores will convert to a letter grade as follows:

• 93.0 - 100 A
• 90.0 – 92.9 A-
• 87.0 – 89.9 B+
• 83.0 – 86.9 B
• 80.0 – 82.9 B-

with the pattern repeating for C and D grades; 59.9 or less is a failing grade.

Classes requiring rescheduling

Due to my travel in the Fall, 2012 semester, I will not be able to deliver the lectures on Oct 11 and Dec 11. These will either be taught by TAs, or will be rescheduled. More information will be provided at the appropriate time.

Honor code

The Emory College Honor Code applies to all homework assignments.

Topics to be covered

This course will emphasize that all living systems have evolved to perform certain tasks in specific contexts. There are a lot fewer tasks and contexts than there are different biological solutions that the nature has created. The problems, which live on the intersection of physics and biology, are universal, while the solutions may be organism– specific. Focusing on physics-style mathematical models of biological processes allows us to uncover phenomena that generalize across different living organisms – something that traditional empirical approaches cannot do alone.  This course will try to take this point of view while analyzing what it takes to perform one of the most common, universal functions performed by organisms at all levels of organization: signal or information processing and shaping of a response (variously known as learning from observations, signal transduction, regulation, sensing, adaptation, etc.) Studying these types of phenomena poses a series of well-defined questions: How can organisms deal with noise, whether extrinsic, or generated by intrinsic fluctuations within them? How can organisms ensure that the information is processed fast enough for the formed response to stay relevant in the ever-changing world? How should the information processing strategies change when the properties of the environment surrounding the organism change? These biological questions are, in fact, physics problems. Equally importantly, they are problems that should be studied in the language of mathematics.

We will study these questions focusing on specific biological examples, including, in particular, bacterial chemotaxis, vertebrate vision, neural computation in insect and mammalian brains, adaptation in bacterial populations, and certain behaviors of rodents.

Tentative class schedule

The schedule of topics covered during each lecture is subject to change. I will revise it periodically to reflect the pace of the class. The current schedule can be found on the class web site.

Week 0

Introduction

1. Why study information processing in biology?
2. Introduction to the class structure and topics
3. What are the problems that organisms need to solve to survive in a changing environment?

Week 1-6

Randomness in biology

1. E. coli chemotaxis as a motivating example for introduction of concepts from probability theory.
2. Basics of probability theory.
3. Inference: organisms responding to their experiences use the same mathematical tools as we use to learn from experiments.
4. Randomness and bacterial survival: Darwin vs. Lamarck.
5. Central limit theorem and the beauty of Gaussian random variables.
6. Random walks and diffusion in different dimensions and the search for transcription factor binding sites.
7. Random walks and diffusion in development – the diffusion equation approach.
8. E. coli gene expression as a model for stochastic chemical kinetics, Master, Fokker-Planck, and Langevin approaches to modeling.
9. mRNA to protein noise propagation in E. coli gene expression.

Week 7-9

Quantifying the fidelity of information processing in biology

1. Bacterial gene expression and protein signaling as examples of information channels – introduction to information theory.
2. Is one bit a lot? What are the limits on information processing imposed by structure of biological systems, temporal structure of signals, and their intrinsic randomness?
3. Does biology care about bits? Information theory and bet hedging in population biology.
4. Projects will be assigned around week 7.
5. Midterm will follow this block of the class, tentatively October 30.

Week 10-14

Time is of essence -- Dynamical information processing, noise filtering, and adaptation

1. Vertebrate vision: linear response approaches to analysis of propagation of fluctuations through biological network. Biochemical enzymatic amplifiers.
2. Spatiotemporal averaging in vertebrate cones as a tool for noise control.
3. Waiting as noise suppression – vertebrate rods quantum bumps.
4. Thresholding and positive feedback as a noise suppression tool. Introduction of dynamical systems approaches.
5. Multistability in protein signaling and gene regulation.
6. Collective behavior as noise suppression – from neurons to immune system.
7. Neural computation in fly vision: adaptation maximizes information transmission.
8. Flashback to chemotaxis and vertebrate vision – negative feedback as a mechanism for adaptation.
9. How fast can adaptation happen? Rats, birds, and flies as ideal detectors of change.
10. In-class project presentations will be on December 4.