FlyLab Experiments, Set II

Analyze the following traits using the FlyLab. For each of these crosses you will be given two traits to analyze at a time. Thus in addition to determining the inheritance pattern of each trait individually you need to explain the inheritance patterns that occur when both mutations are involved in the same cross (a dihybrid cross). In addition, you should use the chi-square test to test any hypotheses that you propose, if a chi-square test is possible for your hypothesis. Don't forget to give phenotypic and genotypic diagrams of all relevant crosses, with results and your conclusions. You will work in pairs (assigned in class) and just turn in one report per pair. There is a scoring rubric for each of these problems showing the characteristics expected in good write-ups of the problems (it is different than the one for FlyLab I so be sure to check it out). The four problems are worth five points each and your results are due in your discussion section, the week of Feburary 12th. I will quickly walk you through setting up a dihybrid cross and analyzing the results with a chi-square test, or you can skip down to problem 1.

Sample walk -through: For a dihybrid cross we need two different traits to follow. Let's try shaven bristles and brown eyes.

Step 1, designing the flies:

  1. Log in to the FlyLab site
  2. Click on the button that says "Start FlyLab" and wait for the program to download to your computer (this may take a couple of minutes). When finished downloading you should see a a gray drawing of two fruit flies.
  3. As in VF 1, design a female fly with shaven bristles and a male fly with brown eyes and then mate them.
  4. You should get an F1 generation that is all wild-type (wt), suggesting that both of these traits are recessive. Now cross the wt F1 to produce the F2 generation.
  5. When the results come back you will see two wt progeny and an arrow pointing down, below the statement that there are "six fly images". If you look at the other images you will see flys with shaven bristles, like the parental female, flies with brown eyes, like the parental male, and at the bottom, flies with both shaven bristles and brown eyes. This is different than the result s we've seen before because we have a new fly phenotype that doesn't look like either of the parents or a wt.
  6. Click on "Analyze Results" to see the numbers for these different types of progeny.
  7. There's lots of different numbers here but if you look close you'll see that there is little difference between the numbers for males and females so select the radio button beside "Ignore sex".
  8. Now there should be just four sets of numbers, for wild-type, shown as a +, shaven (SV), brown (BW) and shaven and brown together, (SV,BW). The numbers will seem a little strange, but there should be much more wt than anything else and fewer SV, BW than the others.

Step 2, determing the expected ratios:

  1. If the traits are assorting independently (as Mendel's pea traits did) than we would expect 3/4 of the F2 to have wt bristles and 3/4 to have wt eyes. If you add together the + flies and the BW flies (which have wt eyes) the number should be about 3/4 of the total (if you used the default this will be about 750/1000).
  2. Probability theory says that if 3/4 of the flies have wt bristles and 3/4 of the flies have wt eyes and these two traits are independent, then 3/4 x 3/4 = 9/16 of the flies should be wt for both bristles and eyes.
  3. As 1/4 of the flies should have shaven bristles, then we would expect 1/4 x 3/4 (wt eyes) = 3/16 SV
  4. The same type of calculation gives us 3/16 BW and 1/4 x 1/4 = 1/16 SV,BW. It may be hard to look at your numbers and see if you really got those ratios, especially as there can be a lot of variation in the results.
  5. To see if your numbers really match what we would expect for independent recessive traits, we need to do a statistical test called a chi-square test.

step 3, doing a Chi-Square test:

  1. Click on the tab labeled "Chi-Square Analysis" near the top right corner of the results page.
  2. In the new window that appears we have the phenotypes and numbers from before along with a column of empty fields.
  3. First click on "Ignore Sex".
  4. To test our results we need to put in the expected ratios in the empty fields. As we expected the numbers of different fly types, as calculated above, to be in a ratio of 9 + : 3 SV:3 BW: 1 SV, BW enter a 9 opposite the +, a 3 opposite the SV, a 3 opposite the BW and a 1 opposite the SV, BW (you could also put in 9/16, 3/16, 3/16, and 1/16, but this is easier).
  5. Now click on "Test Hypothesis".
  6. You should get back a table with the Chi-Square Term for each row and a total "Chi-Square Test Statistic" at the bottom along with the degrees of freedom, which should be 3, and the probability of getting a Chi-Square Test Statistic as large as the one you got if your hypothesis was true, the "Level of Significance".
  7. For scientific purposes, a level of significance above .05 is acceptable and the program will give a recommendation of "Do not reject your hypothesis".
  8. Most likely you have a chi-square score below 11 and a recommendation to not reject your hypothesis.
  9. However, 1 time out of 20 the test will reject your hypothesis, even if it is true.
  10. If this happens to you just ned to repeat the cross and try the test again.
  11. If the traits are not independent or the traits are not simple recessives with 3:1 ratios, then it will fail this test again and you will need to come up with a new hypothesis. This will almost certainly happen in some of the following problems, so beware.

For each of the following problems you should determine the inheritance pattern of each of the two traits, make a prediction for what will happen in a dihybrid cross, and test your prediction with a chi-square test. If the test fails then you will need to come up with a new hypothesis to explain your results, and , if possible, test that hypothesis. You should do both an F2 cross (cross two F1 flies) and a test cross of a heterozygous F1. If you're getting confused, you can always cross just one trait at a time, as in the first assignment.

1. Investigate the radius incompletus wing and purple eye traits (5 pts.)

2. Investigate the ebony body and curly wing traits (5 pts.)

3. Investigate the eyeless and sepia eye traits. (5 pts.)

4. Investigate the black body and apterous wing traits (5 pts.)

Does it matter whether you start with a cross between black and apterous or between black, apterous and wild-type?

Bell CSU Chico Library
This document is copyright of Jeff Bell
Last Update: Friday, January 26, 2001