Quantcast

«

»

Apr 08 2014

It worked!

At last I’m home after a long day, but I have good news: my genetics students have been doing a three point mapping cross for the last month or so, which is always worrisome because if they screw it up, you don’t know the results until the very end of the experiment. But now the data is trickling in, and with a quarter of the numbers done, it’s looking really good.

It’s a big chunk of time and effort, and it feels awful if it flops at the last analysis.

9 comments

Skip to comment form

  1. 1
    Al Dente

    my genetics students have been doing a three point mapping cross

    Um…that’s nice…I guess..er, I’ve got one small question. What’s a three point mapping cross?

  2. 2
    JohnnieCanuck

    Al @ 1.

    I looked it up on Wikipedia. That’ll get you started and may be all you ever wanted to know, but it looks like that page needs a little more work. It doesn’t even have references. There are lots of other results on Google for those who want to go into more depth.

  3. 3
    Joseph Felsenstein

    Back in the 1970s I used to teach an elementary biology course at our university. We had a three-point cross (a cross with three marker genes) of Drosophila. I reworked the exercise down to two loci, so a two-point cross. The students had to count the numbers of four genotypes, involving the two possible phenotypes at an eye color gene and the two possible phenotypes at a wing shape gene.

    You could always tell which students had cheated and “dry-labbed” the exercise, because they got all four possibilities roughly equal in numbers. If you actually counted, the two recombinant classes were less frequent. I’d say about 20% of the students were dry-labbing.

  4. 4
    violetknight

    Genetics was the last really fun course in the bio major (at least at my university).

  5. 5
    ChasCPeterson

    Joe Felsenstein: Not clear on the specifics, but if by “dry-labbing’ you mean falsifying data in lab reports, then 20% represents an abject failure in science education. But of course if laboratory exercises are always traditional rote ‘cookbook’ labs, with a right answer (i.e. it ‘worked’ or it ‘didn’t work’) then of course busy lazy immature students are going to make up the ‘tight’ data. What is actually learned when that happens?

  6. 6
    rq

    Sounds intense, nerve-wracking, and fun, in that slow biology way. Yay for successful results!
    (Never collaborate, as a microbiologist, with physicists. Their sense of time is completely off. “I plated yesterday, sir, I can’t count colonies until tomorrow!” “But but but… why??”)

  7. 7
    PZ Myers

    I’m not seeing any signs of fudging in the student data here — they were actually rather horrified and mystified when they presented their individual counts in lab yesterday, and they’d seen things like disparities in the sexes (‘I only saw 30 males & 120 females — did I do something wrong?’), and then when we added everything up, all those little vagaries of chance started to even out, and the numbers started to look like they fit together and we could actually make a map.

    Most interesting, though, is that they’ve got this big pile of data for 8 different phenotypes in two sexes, and I tell them that what they have to do is analyze it creatively — are there patterns? Is one particular allele more deleterious than the others? Is one sex more robust than the other? Etc. I want them to come up with hypotheses and also propose experiments to test them.

    I also have them do a two point cross earlier in the semester. That one did not work out at all well — the numbers are all over the place, and it was clearly a combination of students who were new to this and had no idea what they were doing, and using an allele that was an eye color variant that was way too subtle for students who were just figuring out which end of the fly was the head, and which was the butt. But it turned out to be good training — by the time we got to the three point cross, they were pros at breeding and scoring flies.

  8. 8
    moarscienceplz

    I’d never heard of a three point mapping cross before, and that wiki article is virtually useless, at least for a noob like me, but by poring over it, I think I at least have a metaphor for this process (please do tell me where I go wrong):
    You have a long, skinny, half-loaf of garlic bread. It was made by a quirky baker who added a slice of mushroom, a slice of olive, and a basil leaf, but you are blindfolded and they can be anywhere on the loaf. The loaf is a chromosome and the three weird ingredients are three dominant genes you would like to locate, or at least learn something about their locations.
    You ask the weird baker to make 1000 more identical loafs of garlic bread, then you slice them (while still blindfolded) and then pass out the pieces to a large group of hungry undergraduates, asking them to report what’s on their piece before they nom it.
    (This metaphor fails here, because it would imply a much larger number of pieces would have no weird ingredients than what would happen in the real experiment, but please ignore that.)
    If a reports come back of pieces with all three items, then that tells you that the items are pretty close together. Further, if you get some reports of basil and olive without mushroom, and other reports of basil and mushroom without olive, then you can deduce that the basil is placed between the olive and the mushroom.

    So, how’s that?

  9. 9
    Rich Woods

    When I got home from work this evening I just wanted to relax, but instead I found myself learning something.

    Results and conclusions, please, PZ. Don’t keep us on tenterhooks. Feel free to whip your students if you have to.

Comments have been disabled.