Value in performing whole genome sequencing on sample size of 1

Question

[ehzed]

All the papers that I've linked below, they performed SNP calling on sequenced reads from one animal for each breed (e.g. sequenced only one Black Angus Bull or one male broiler chicken): http://www.nature.com/nature/journal/v432/n7018/full/nature03156.html

http://bmcgenomics.biomedcentral.com/articles/10.1186/1471-2164-12-559

http://gbe.oxfordjournals.org/content/5/7/1376

My question is, are they assuming that every animal belonging to that breed is completely identical? What if a snp is only unique to that one animal specifically? What is the value in performing snp calling on individual animals and not pooled samples? Thanks

Answer 1

[Brian Bushnell]

The majority of SNPs not seen in other breeds will be specific to the breed rather than the individual, because breeds are inbred and they have not had very long for genetic drift within the breed. There will be some that are specific to an individual/herd/flock, and those will be indistinguishable, and some will be missed because they are not homozygous in the breed, but that's the tradeoff for shallow sampling.

[ehzed]

Hi Brian, can you elaborate on that? If they only sequence one animal from each breed, they just assume that the snp is found in all animals in that breed, and that the snp didn't arise spontaneously in that one animal? And their only argument for that assumption is that most domestic breeds are inbred? Are there livestock sequencing studies that measured variability within breeds that I can read that confirms that we can make these kinds of assumptions? Thank you!

[Brian Bushnell]

Hmm, I should clarify that I have not seen any data on livestock; those are my assumptions based on working with human racial SNP frequencies (from dbSNP, HapMap, etc). Generally, when you sequence an individual human, virtually all of the SNPs have been observed elsewhere in other people. If you sequence an individual of a certain race, then, you can roughly categorize their variations into 3 groups:

1) Those present at some level in most or all human populations.

2) Those exclusive to that race or ancestral branch (for example, any Native American race that went through the same bottleneck event, presumably from Asia).

3) Those exclusive to the individual/family/village.

Category 1 is much larger than category 2, and category 2 is much larger than category 3; and this is for people, who have generally be able to breed freely, and have had much more time to accumulate new variations compared to livestock. So since category 3 variations are quite rare in people (a lot of research depends on this fact), they should be even more rare in inbred livestock.

[Brian Bushnell]

Also, even if fully half of the new SNPS detected in, say, Black Angus were specific to the individual, it would still be useful. Let's say you wanted to design a SNP array to test rapidly and inexpensively whether a sample of beef was Black Angus. First, you make a SNP array based on your single individual. Then, you calibrate it by testing on 100 diverse cattle that are all Black Angus, and see which SNPs they all share. Then, you simply ignore the signals from the unshared half of the SNPs in the array and use the remaining half as a cheap, high-precision test for Black Angus beef.

[GenoMax]

This article is enlightening. A single bull is responsible for a large % of DNA in all holstein cows today. This bull also had a mutation that led to spontaneous abortions in some of the calves but still the overall positive impact on the dairy industry has been in billions of dollars.