Statistics for Molecular Genetics

Statistics/Genetics 537

Fall, 2002

News

• 2003-01-07: Final grades now include project 3 and final exam grades. If you want your final exam or project 3, let me know.
• 2002-12-24: Final grades.
• 2002-12-21: Detailed answers to final.
• 2002-12-16 (10:19am) : Useful collection of links for final preparation (All these documents are linked elsewhere on this page, but are collected here in one place):
  • Homework 1 questions
  • Homework 1 some solutions
  • Homework 2 questions
  • Homework 3 questions
  • Project 1
  • Project 1 addendum question about EM
  • Project 2
  • Project 2 answers, Homework 1 answers (labeling corrected)
  • Homework 3 answer, EM answer (typo in &mu_Bb expression corrected)
• [new]2002-12-16 (10:16am) : Remember, there are some student-contributed answers for homework 1, but not nearly all of them. Other homework 1 answers are in the reviews (next two notes).
• 2002-12-12 (10:00pm) : Notes from review session 2 posted. If you still have questions, there is email.
• 2002-12-11 (8:41pm) : Notes from review session 1 posted and updated. I made a algebra mistake in discussing the solution to 2(d) so the upper bound is 0.28, still inclusive of 0.25.
• 2002-12-11 (1:11pm) : Compute and verify the answers plus more for your project 1 data set. Beware, it is a little slow.
• 2002-12-11 (1:11pm) : Need to logout so someone else can use this computer to retrieve their data? Try here or here. If it asks you for a password, then you have been successfully logged out.
• 2002-12-05 (7:58pm) : Lynch and Walsh partial chapter. IMPROVED VERSION. If you had trouble viewing this file before, it should now work.
• 2002-12-02 (2:31pm) : If you are having difficulties calculating log likelihoods for homework 3, problem 1, please be sure to write the complete likelihoods and the details of the statistical test, even if you can't calculate a statistic yourself. Explain what happens when you try. Use the following likelihood results to calculate your statistic, or if you can calculate likelihoods, compare your answers to

  Maximum likelihood under Ha	8.147
  Maximum likelihood under H0	2.303

• 2002-11-16 (6:38pm) : Answers are now posted in review 2. Got answers? We need to fill in the details of the EM algorithm that would be used to estimate segregation distortion in the data from project 1. As mentioned in lecture 21, you should be familiar with this approach as it could appear on the final.
• 2002-11-16 (5:25pm) : I have fixed the fluff in project 2. Please read the new version posted below. The data and analysis are unchanged, but questions for discussion have been clarified with more hints/information added to help. Ask me if there are other questions. Thanks to those who have come with questions. And thanks for your patience as I work out the bugs.
• 2002-11-15 (11:00pm): Homework 3, problem (1) has been rewritten yet again to provide a few more details. Also, the mle is 0.045, not 0.43. New pdf is updated. Sorry, sorry.
• 2002-11-15: Project 2. While you should be able to perform the statistical tests, the fluff and questions in the description of this project are more than a little rough. Please do the statistical analysis, but check here for help with any discussion/interpretation. I am working on fixes.
• 2002-11-15: Homework 3, problem (1) was rewritten for clarification. The rewrite also corrected the invalid statement that P(M_iD) = P(M_i | D).
• 2002-11-13: Project 2. Ignore the parts about estimating recombination fraction θ for the affected sib pair data. Wrong thinking behind that question. I'll try to clarify my point on that dataset later. You should be able to do the analysis as in the Sham example, even if you can't estimate a parameter.
• 2002-10-21: Due to the many errors in lecture 11, the homework due date is postponed to next Tuesday for onsite students.
• 2002-10-21: Lecture 11, many problems with log likelihoods: a lengthier correction has been posted.
• 2002-10-18: Lecture 11, slide 9 correction: there should be no 2 in the first log in that log-likelihood function. This is relevant for homework #2.
• 2002-10-16: Final exam will consist of modified homework problems. We will work together to build a solid answer key to the homework problems.
• When an onsite due date falls on a day when there is no class, you may turn in the homework to either of my offices (under the door if I'm not there) or via email.

Assignments

• Due Dec 20, 2002: Project 3: This is called a project, but you need not write this up as a scientific manuscript. Just answer the questions.
• Due Dec 2, 2002 for everyone: Homework 3 (I highly recommend campus students turn this homework in before Thanksgiving.)
• Due Nov 22, 2002 (Mayo: Dec 2, 2002): Project 2: This project has two parts. Forget your password?
• Due Oct 29, 2002 (Mayo: Oct 31, 2002): Homework 2: 3-locus order assignment
• Due Oct 25, 2002 (Mayo: Nov 1, 2002): Project 1 (If you do not already have an account, please make one or if you forgot your password, please use the retrieval form). Answers.
• Due Sep 27, 2002 (Mayo: Oct 11, 2002): Homework 1: Segregation ratio assignment
• Week of 2002-09-16: Read Sham, Chapter 3
• 2002-09-05: Read Sham, Chapter 2 (excluding 2.6 & 2.7)

Lectures

30. 2002-11-12: Review II
29. 2002-11-10: Review I
28. Guest lecturer on variance components QTL mapping
27. Guest lecturer on QTL interval mapping
26. 2002-11-21: Quantitative Traits VI: Parent/Offspring, Sibs, Twins
25. 2002-11-19: Quantitative Traits V: Environmental effects and resemblance of relatives
24. 2002-11-14: Quantitative Traits IV: Epistatic effects and quantitative traits
23. 2002-11-12: Quantitative Traits III: Single-locus QTL for backcross and F2
22. 2002-11-07: Quantitative Traits II: Note, the dataset I used to discuss project 1 did not have a total of 30 offspring (for some unknown reason), but when I calculated the segregation proportions, I assumed 30. In any case, the concepts should be correct, as the relative proportions remain far from 1:2.
21. 2002-11-05: Quantitative Traits I
20. 2002-10-31: Association Studies III
19. 2002-10-29: Association Studies II
18. 2002-10-24: Association Studies I
17. 2002-10-22: Non-Parametric Methods II (also Model-Free Linkage Analysis II)
16. 2002-10-17: Non-Parametric Method I: Allele Sharing (mislabeled lecture 17)
15. 2002-10-15: Linkage Analysis VII: solution to ascertainment model discussed
14. 2002-10-10: Trapped in the airport; no lecture.
13. 2002-10-08: Linkage Analysis VI: corrections and minor clarification with posting; some slides not covered
12. 2002-10-03: Linkage Analysis V
11. 2002-10-01: Linkage Analysis IV
10. 2002-09-26: Linkage Analysis III
9. 2002-09-24: Linkage Analysis II: Unknown or Mixed Phase
   • Again, this lecture comes with a problem to be worked in class. Answers are not yet here, though I hope to find some time to put them online.
   • The problem to do is: Find a maximum likelihood estimate for the frequency u of allele A at a single locus for a codominant and dominant marker (two groups, one does codominant, one does dominant). The experimental data available to you are the counts of AA, Aa, and aa offspring from a maternal heterozygous (Aa) plant: n_AA, n_Aa, and n_aa, though for the dominant case you only have two counts n_A-=n_AA+n_Aa and n_aa. You must assume the outcross rate t is known.
8. 2002-09-19: Linkage Analysis I: Known Phase Problems
   • Note for Mayo students. This lecture involves a problem to be worked in class. The answers are on the slide; it would be best to work the problems before looking at the answers.
   • Also, for everyone, see new material in these slides that will help explain the ones I didn't get to and will improve my screwed up explanation of the conditional probability calculation.
7. 2002-09-17: A First Look at Linkage Analysis
6. 2002-09-12: Segregation Analysis II
5. 2002-09-10: Segregation Analysis I
4. 2002-09-05: Statistics Review II
3. 2002-09-03: Statistics Review I: Note there is a mistake in the Mendel example for conditional likelihood
2. 2002-08-29: Biology Review II
1. 2002-08-27: Biology Review I (make account first)

Resources

• Allele-Sharing Linkage Analysis Bibliography
• Imprinted Gene Catalog; more on gene imprinting
• How to review a linkage paper
• History of the Elston-Stewart algorithm from Stewart.
• A break from seriousness: Statistical Genetics Humor Page
• Learn the background genetics by reading the first three chapters of An Introduction to Genetic Analysis.
• Movies of mitosis and meiosis at the The Biology Project, developed at the University of Arizona.
• If you ever run into problems with genetics vocabulary, you might want to try Biotech's Life Science Dictionary.

Last update: Thu Dec 12 21:48:49 CST 2002