Administration
When: Wednesday, 6:10pm to 8pm ; Fall 2006
Where: 407 MATH 337 MUDD
By who: Itsik Pe'er, office hours: Wed 5-6
One time change: Nov 1st: Wed 4:30-5:30
Teaching assistant: Nalini Kartha, office hours: Tue 2-4, MUDD 122A
What for: 3 credit points
For who: Graduate students of relevant fields.
Although the course is listed in Computer Science,
cross enrollment by students with biomedical background is encouraged.
Abstract
This course is intended to introduce students of both computational
and bio-medical skill sets to current quantitative understanding
of human genetics and prepare them to computational research in the field.
Topics include: genetics of a single site, coalescence with recombination,
history of humans, mapping rare mutations through linkage,
mapping common variants through association,
isolated and admixed populations, natural selection, copy number changes,
model organisms, and genotyping technologies.
The computational toolbox discussed includes parameter inference,
likelihood analysis, hidden Markov and other graphical models,
eigenvalue decompositions, and classification problems.
Tentative Program
- Sep 6 Intruction to genetics[ppt][pdf]
- Sep 13 Genetics of a single sitea [ppt][pdf]
- Sep 20 Coalescence with recombination[ppt][pdf]
- Sep 27 A brief history of humans[ppt][pdf]
- Oct 4 Mapping rare mutations: Linkage[ppt][pdf]
- Oct 11 Technical class: Hidden Markov Models (Nalini Kartha)
- Oct 18 Common variants: Association
[ppt][pdf]
- Oct 25 Isolated and admixed populations
[ppt][pdf]
- Nov 1 Natural selection
[ppt][pdf]
- Nov 8 Copy number changes
[ppt][pdf]
- Nov 15 Genotyping technologies; Non Humans I
[ppt][pdf]
- Nov 22 Project outline presentations
[ppt][pdf]
- Nov 29 Non humans II + project technical advisory
- Dec 6 Project final presenteations + Discussion
FAQ
- Q:
I'm a Computer Science student with no background in biology.
Can I take the course?
A:
Sure. The necessary biological background will be given in condensed
form at the first meeting, and then supplemented when necessary.
The course is designed for a small, interactive class, with each student
having enough weight to affect the level of background given.
- Q:
I'm a student in Medical Informatics/Biological Sciences with no background in Computer Scoience.
Can I take the course?
A:
Sure.
You will need to accept some leaps of faith during classes,
and you need some background in probability/statistics/biometry.
The course is designed for a small, interactive class, with each student
having enough weight to affect the level of background given.
- Q:
How is the course graded?
A:
Ideally, I'd like you to do final projects in pairs.
Each pair will include a student with primariliy engineering background,
and a student with primarily biology background.
This is possible in an intimate class setting.
If enrollment surprises me for the better,
or if the class is out of balance between scientific disciplines,
this format will be reassessed.
- Q:
What are the projects like?
A:
An example may be giving you a 5 year-old paper, asking you to read, understand,
implement the computational method so that it can work on today's data
(which is often orders of magnitude larger and more complex),
analyse what you find, and write up what you find.
Submission will include:
Any code you've developed
- A detailed written report
- An executive summary of 5 slides.
- Q:
What are the office hours?
A:
Just before class. See my contacts.
- Q:
Is there a book with the course material?
A:
Sorry, but no. The course is given from the first time,
and material is being picked up here and there,
much of it from the recent scientific literature.
The slides for each lecture will include references.
- Q:
I can't find copies of The Handbook of Statistical Genetics and Coanescent Theory
in the library.
A:
Try a bigger library.
Relevant chapters are online for
The Handbook
and Coalescent Theory