EEB 5348 -- Population Genetics

Berger paper

2010-12-12T11:38:33Z

I apologize for not getting to this yesterday, but I just checked the Berger et al. (2010) reference, and it seems to be fine. I used the DOI and got right to the page, but here's the direct link: http://www.pnas.org/content/107/41/E157.

In case you don't know what DOIs are or how to use them, here's a brief explanation:

"DOI" stands for "digital object identifier", a system

for identifying content objects in the digital environment. DOI^® names are assigned to any entity for use on digital networks. They are used to provide current information, including where they (or information about them) can be found on the Internet. Information about a digital object may change over time, including where to find it, but its DOI name will not change. (from http://www.doi.org/)

The Berger et al. reference looks like this in the notes:

James Berger, Stephen Fienberg, Adrian Raftery, and Christian Robert.
Incoherent phylogeographic inference.
Proceedings of the National Academy of Sciences USA, 107(41):E157-E157, 2010.
10.1073/pnas.1008762107.

That last line is a DOI. To use it to get the paper simply put type the following URL into the address bar of your favorite browser:

http://dx.doi.org/10.1073/pnas.1008762107

http://dx.doi.org is the URL of the database that keeps track of DOIs. To find a specific DOI, you simply type the DOI at the end of that URL (after a "/") and it will take you there.

Final set of notes posted

2010-12-08T21:00:29Z

I made the final set of changes to my notes on approximate Bayesian computation and uploaded them a few minutes ago. You'll find them in the link for today's lecture. I'll also add them to the outline of notes as soon as I finish typing this entry. Now that I've finished them, it seems pretty clear that I won't have more than a few minutes to say anything about next generation sequencing, but I've added links to several papers where you can find more information on what sorts of things may be possible and the challenges to making those possibilities real.

On a different note, please remember that you should send me your answer to Problem #7 no later than 5:00pm next Wednesday, 15 December. I guarantee that I'll get every paper graded, meaning I can assign final letter grades, if I get your paper by then. If I don't get it by then, I might be able to get it graded in time to turn in something other than an incomplete, but I can't make any promises.

Study Sessions for Problem 5

2010-11-08T20:20:26Z

Hi all,

I'm waiting for final confirmation but tentatively the study sessions will be the following:

Wednesday Nov 10 from 1-5

Tuesday Nov 16 from 11-2

I don't have a lot of free time other than those slots, so PLEASE try to bring your questions to at least one of them!

Cheers

Kathryn

UPDATE: I have a doc apt on Wed, so I will be gone from 2-3. Sorry!

Expectation for Friday Class

2010-11-04T19:42:59Z

Hi all,

The goal for Friday's class is to make sure everyone knows the following:

1) How to run R

2) How to get the output of R into a format you can actually use

3) How to use GenBank

In order to accomplish this goal, please download R, the data file and the "key to SNPs" file before coming to class. Saving all of these on your desktop will also make life easier.

If you already know how achieve the above mentioned goals, feel free to take the day off.

One final thing: Please think about when you would like me to hold a study session for this problem. I'd really like to only hold one!

Cheers

Kathryn

Preview of association mapping

2010-10-28T13:37:45Z

The issue of Nature that just came out happens to have a bunch of articles relevant to association mapping, which we'll start talking about on Monday. The report on the 1000 genomes project, the accompanying review by Rasmus Nielsen, and the overview of human association mapping by Monya Baker are particularly relevant. If you have a chance, you may find it helpful to glance at them before Monday.

Solution to Problem #3

2010-10-25T16:54:57Z

Kathryn typed up her answer to Problem #3. I've converted it to a PDF and posted it. Feel free to contact either of us with questions about it.

Expectations for Problem 4

2010-10-24T16:15:11Z

Hello everyone!

Just so that we are all clear, since Problem 4 is a "larger" problem, I am expecting essentially "results" paragraphs for questions 1 & 2 and a page-long "discussion" for question 3, as well as the code for solving the problem.

I'd recommend reading the paper associated with the problem if you haven't already!

Cheers

Kathryn

Exponents in WinBugs/OpenBugs

2010-10-21T17:48:08Z

Hi all,

I've gotten a couple of questions so I thought I'd let you all know how to code for an exponent in WinBugs/OpenBugs.

Let's say you want to raise 10 to the power of 2 (ie 10 squared).

Rather than using 10^2 which is what Excel has drilled into our brains, you type

pow(10, 2)

where pow is the "power" function, 10 is the factor and 2 is the exponent.

If we change the situation and decide that we want to raise (1-x) to the power of 2 we can do this 2 ways:

pow ((1-x), 2)

y<- 1-x

pow (y, 2)

However, WinBugs/OpenBugs won't actually let you do a power function the first way, so you'll have to use the second notation. Remember that x and y can be anything you want them to be (since they are just symbols) so make sure they match the rest of your code!

Hope this helps!

Kathryn

2010-10-20T10:44:19Z

You might be interested to know that today is World Statistics Day. Just what you wanted, right? A whole day to celebrate statistics.

Updates

2010-10-18T23:10:02Z

I've uploaded a new version of Problem #4 It includes the two corrections I mentioned in lecture this morning.

I've corrected the formula for Var(p_t).
I've corrected a couple of entries in the table of data.

I've also uploaded a commentary on Problem #2 that Kathryn put together.

Study sessions for Problem 4

2010-10-13T13:01:53Z

Hi all,

The usual place: Pharm/Bio 303

Wed Oct 20 1-4

Tues Oct 26 11-2

My schedule is reasonably open, so let me know if you need to meet at another time.

Kathryn

More on effective population size

2010-10-06T12:40:07Z

It just dawned on me that I forgot to mention that Charlesworth and Charlesworth have a good discussion of effective population size in 5.2. By "good" I don't mean that it's easy to follow. It's not. But it's reasonably comprehensive, it's accurate, and it has some good examples.

Study sessions for Problem 3

2010-09-29T19:09:36Z

I've booked the Pharm/Bio fishbowl (room 303) for the following dates and times. If you know you need other time, please let me know.

Tuesday October 5: 11am to 2pm

Wednesday October 6: 1:15pm to 4pm

Cheers

Kathryn

Justifying your K value for Problem 2

2010-09-28T15:15:11Z

Hi all,

When you are answering question 1, please make sure to include the bar graphs from Structure for your chosen K (and perhaps a few other K's as well). One thing that I have noticed is that different iterations result in different clustering, so if I don't know the diagram you are basing your cluster on, I'm going to have a hard time grading!

Also, remember to send me your file with the following name format: LastName_Problem2

Cheers

Kathryn

More on picking the "right" K from Structure

2010-09-25T16:47:22Z

I got an e-mail this morning with a question that deserves further discussion, so I'm posting the question and my response here so that everyone can see it.

Question: After running Structure last week, we all found that as K got larger, so did its corresponding mean log prob of the data. That would mean that K=8 has the largest and so, logically, we should pick K=8 for our K values to run with the new data? I remember you talking about this in class and I can't find in my notes if this is the most important method for finding K in our situation.

My response: When we first talked about Structure in class, I presented two methods for selecting K: (1) looking at the log probability of the data directly from Structure (in our case a mean of ~20 runs for each K from the spreadsheet) and (2) calculating Delta-K. I pointed out that simply using the log probability of the data is likely to overestimate K (a) if the underlying groups have genotype frequencies that depart from Hardy-Weinberg &/or (b) if the multilocus genotype frequencies aren't simply the product of the single-locus frequencies. In our case both (a) and (b) are likely to hold, so basing our choice of K purely on the log probability of the data probably isn't a good idea.

Looking at Delta-K isn't a horrible alternative, but if you read the paper in which it was proposed as an alternative, this approach to selecting K was tested only under one very specific kind of population structure, which may or may not apply in our case. So I'd certainly suggest looking at Delta-K, but I wouldn't treat what it tells you as gospel either.

So what do you do? Well, take a look at what we did in our Molecular Ecology paper. We combined what both criteria had to tell us about K with a look at where populations fell on the map and what the different choices of K seemed to tell us about their relationship. In other words, we combined some quantitative understanding of population structure with biological intuition about what relationships are likely to be important to settle on a K that best helped us to understand the data.

That's what I'm suggesting you do here. Look at both the log probability of the data and Delta-K. Then combine what they tell you with what the geographical distribution of populations might suggest about relationships to select a K (or possibly 2-3 Ks) that seem most informative. Be sure to explain your reasoning for settling on the number of Ks that you choose. Then use that choice (or those choices) to set up the analyses for questions 2 and 3.