Goodness-of-fit chi-
square test
Mendelian Genetics
• Law of Random Segregation: alleles segregate randomly in
the formation of gametes
• Law of Independent Assortment: unlinked traits are
transmitted to offspring independently of one another
Chance and Independent Assortment
Monohybrid Dihybrid
Mendelian ratios Non-Mendelian ratios
Evaluating the Goodness of Fit— χ2
Analysis
• An important question to answer in any genetic experiment is how can we decide if
our data fits any of the Mendelian ratios we have discussed.
• Punnett Squares are convenient for predicting the outcome of monohybrid or dihybrid
crosses.
• The expectation of two heterozygous parents is 3:1 in a monohybrid cross or 9:3:3:1 in
a dihybrid cross.
• Performing a three or four trait cross becomes very messy. In these instances, it is
better to follow the rules of probability.
• Probability is the chance that an event will occur expressed as a fraction or percentage.
In the case of a monohybrid cross, 3:1 ratio means that there is a 3/4 (0.75) chance of
the dominant phenotype with a 1/4 (0.25) chance of a recessive phenotype.
• A statistical test that can test out ratios is the Chi-Square or Goodness of Fit test.
Evaluating the Goodness of Fit— χ2
Analysis
Example: Coin flipping
Coin #1: 505 heads, 495 tails
Coin #2: 460 heads, 540 tails
Is there something wrong with either coin? How to decide? How much
deviation from the expected values do we tolerate?
Statistical question to ask: What are the chances that even a normal
coin would give a result this far off from the expected result?
The χ2 statistic is used in genetics to illustrate if
there are deviations from the expected
outcomes of the alleles in a population.
Null there are no significant deviations
between the measured results and the
hypothesis
predicted ones
Evaluating the Goodness of Fit— χ2
Analysis
• A measure of how well observed data conform to a specified, expected, or
theoretical probability distribution.
315 yellow, round seed
108 green, round seed
101 yellow, wrinkled seed
32 green, wrinkled seed
556 9.84 : 3.15 : 3.375 : 1
Is this really a 9:3:3:1 ratio? Is the deviation due to chance or is there something
wrong with Mendel’s hypotheses?
χ2 analysis: Test the “null” hypothesis—that the ratio
of
observed offspring is no different from a 9:3:3:1 ratio.
χ2 Analysis of Mendel’s Dihybrid F2
Results
Degrees of freedom (df)
The number of independently varying parameters in the experiment
(number of offspring classes-1) = n-1
Why does df matter?
More independent categories, the more opportunities for chance
deviation; therefore need to make greater allowance for chance
deviation
Look up the P value in a χ2 table
P is the probability that the null hypothesis is true, and a deviation this large is due
to chance
If P >0.05 then we do not reject the null hypothesis
χ2 Analysis of Mendel’s Dihybrid F2
Results
In corn, purple kernels are dominant over yellow kernels, and full kernels are dominant
over shrunken kernels. A corn plant having purple and full kernels is crossed with a plant
having yellow and shrunken kernels, and the following progeny are obtained:
purple, full 112
purple, shrunken 103
yellow, full 91
yellow, shrunken 94
What are the most likely genotypes of the parents and progeny?
Test your genetic hypothesis with a chi-square test.
degrees of freedom = 4 - 1 = 3.
• The calculated chi-square value of 2.7 lies between
2.366 (a probability of 0.5) and 6.251 (a probability of
0.1).
• The probability (P) associated with the calculated chi-
square value is therefore 0.5 < P < 0.1.
• It is the probability that the difference between what
we observed and what we expected is due to chance,
which in this case is high, and so chance is likely
responsible for the deviation.
• We can conclude that the progeny do appear in the 1 :
In goldfish the gold color is believed to be controlled by a single
gene with two alleles, recessive g for gold color and dominant G
for black color. Suppose that in a monohybrid cross generated the
following results. Use Chi-Square to test the hypothesis that the
gold color gene obeys Mendels principle of random segregation?
Two different genes controlling pod morphology in peas are found on
chromosome 5. The narrow pod trait is recessive to normal pods and
yellow pods is recessive to green pods. A heterozygous plant was
testcrossed and the following progeny were obtained.
144 narrow green pods
150 normal yellow pods
110 narrow yellow pods
90 normal green pods
Use Chi-Square analysis to determine if the pod shape and pod color gene
are assorting independently.
