Question

Part 3: Identify Associations Using Real Data (where the associations are not 100%)

When researchers scan the genome using thousands of markers to identify variations that are associated with particular phenotypes, they need techniques to evaluate the strength of the associations so that they know which SNPs are closest to the locations of the genes of interest. You will do a short exercise to help you understand how this is done, followed by a chi-square test on one of the loci.

We will look at coat length

You can find a file containing 12 SNP Cards for dog coat length in this week’s module, under the “lab” heading. Each card represents sequence data obtained from one dog. The sequences show SNPs at seven loci on chromosome 32. The dog DNA was obtained from saliva samples collected by student dog owners. The DNA sequence was then determined by scientists at the Broad Institute in Cambridge.

For each SNP locus on the cards, count the number of times an allele appears on the cards. Record the number in the appropriate box in the tables below. Note that if the SNP at a particular locus is homozygous (e.g., TT), you will count the T allele twice. If it is heterozygous (e.g., TC) you will count the T allele once and the C allele once.

Once you’ve recorded the number of times each allele appears, calculate the difference in alleles between the two groups of dogs and record that number. Add the differences for each allele and record the total number of differences in the last box of each table. The largest total difference indicates the strongest association between that particular SNP and the phenotype.

The first one has been completed for you. Complete the next six on your own (table is on the next slide).

Locus Allele Short coat Long coat 0 chr 32 7420804 I c total differences chr 32 7472206 total differences chr 32 7473337 total differences chr 32 7479580 total differences chr 32 7482867 total differences chr 32 7490570 total differences chr 32 7492364 total differences

Based on your data (table on the previous slide):
1. Which SNP (or SNPs) is most strongly associated with coat length?

2. Which SNP (or SNPs) is least associated with coat length?
Now we are going to use chi-square testing to determine if the differences observed for the locus

identified in question 1, above, are statistically significant differences, or due to random chance. 3. Identify the locus you are testing:
4. Identify the alleles and the number of each in dogs with long and short coats:

5. Calculate the expected number of each allele in each type of dog if there is NO statistical association.

If you need help with this, here is an example:
Suppose you saw 3 C alleles in dogs with short coats and only 1 in dogs with long coats. The total number of C alleles is 4 and if they are evenly distributed between the two traits, we expect 2 of each.
Suppose also you saw 9 T alleles in dogs with short coats and 11 in dogs with long coats. This is a total of 20 alleles, and we expect each type of dog to have 10 if there is no association between the SNP and the trait.

6. Perform the chi-square test below, stating the hypothesis you are testing, do the calculations and draw your conclusion.

Answer 1

1.singlenucleotide polymorphism (SNP) is a DNA sequence variation that occurs commonly within a population. For example, a SNP may replace the nucleotide adenine (A) with the nucleotide cytosine (C) in a particular stretch of DNA. Researchers know the precise location of that SNP in the genome. Once a SNP is correlated with a particular trait (meaning that it occurs more frequently in individuals with that trait), researchers focus on that region of the genome and sequence it. In most cases, the SNP itself does not contribute to the trait, but it is close to the trait-causing stretch of DNA, serving as a marker or signpost for the location. 4. Why do you think it is important to analyze the DNA of many dogs when doing this.SNP can be correlated with a phenotype because it causes that phenotype (e.g., coat length ) or because it is linked to, or located close to, the allele that causes the phenotype.

Hence answer -Chr32 7473337