MathBench > Statistical Tests

Chi-squared Tests

Chi-squared in biology: Testing for a dihybrid ratio

You may have noticed we haven't talked about using chi-squared in biology yet. We're going to do that now.

In biology you can use a chi-squared test when you expect to see a certain pattern or ratio of results. For example:

We'll focus on the last one. If that's what you're also doing in class, what a coincidence!

normal mouse normal mouse normal mouse fuzzy mouse
normal mouse normal mouse normal mouse fuzzy mouse
normal mouse normal mouse normal mouse fuzzy mouse
fanged mouse fanged mouse fanged mouse fuzzy fanged mouse

 

In that module, we considered two (hypothetical!) genes in mice. T and t coded for normal teeth or vampire fangs, respectively, and F and f coded for smooth or fuzzy fur, respectively.

A dihybrid mouse would have one copy of each of the 4 alleles and would look completely normal: normal mouse However, if two such mice mated, they would have offspring that showed the whole range of possible phenotypes: normal ( normal mouse), fanged (fanged mouse), fuzzy (fizzy mouse), and fuzzy and fanged (fuzzy fanged mouse) . Furthermore, these phenotypes should appear in approximately the ratio 9:3:3:1, resulting in a nursery that looks something like this (if 16 babies were born).

But, as we discussed in that module, the process of generating new mice is random, so the ratio will not be exact. Now we have the tools to test whether an actual litter approximates the 9:3:3:1 ratio.

If Mr and Mrs Mouse are both heterozygous for both traits (TtFf), then their offspring should follow the 9:3:3:1 ratio. So if they had 160 babies, how many of each type would you expect?

mouse deer

all normal

fanged

fuzzy

fuzzy and fanged

Mr and Mrs Mouse have 80 normal, 33 fanged, 33 fuzzy, and 14 fuzzy fanged babies. Does this data support the double hybrid model of the data?



observed ("o") expected ("e") (o-e) (o-e)2 (o-e)2/e
normal
fanged
fuzzy
fuzzy fanged
Total 160 160    
Check

     

 

degrees of freedom Χ2 crit for p-value of 0.05
1 3.84
2 5.99
3 7.81
4 9.49
5 11.07
6 12.59
7 14.07
8 15.51
9 16.92
10 18.31

 

Answer: df=3; Chi-squared statistic = 3.11, which is not bigger than 7.81 (the chi-squared critical value), so the data fit the model -- the ratio of phenotypes supports the double hybrid model of the data.