Pespeni MH et al. (2010), Restriction Site Tiling Analysis: accurate disc...

ECB-ART-41603

Genome Biol 2010 Jan 01;114:R44. doi: 10.1186/gb-2010-11-4-r44.

Show Gene links Show Anatomy links

Restriction Site Tiling Analysis: accurate discovery and quantitative genotyping of genome-wide polymorphisms using nucleotide arrays.

Pespeni MH , Oliver TA , Manier MK , Palumbi SR .

???displayArticle.abstract???
High-throughput genotype data can be used to identify genes important for local adaptation in wild populations, phenotypes in lab stocks, or disease-related traits in human medicine. Here we advance microarray-based genotyping for population genomics with Restriction Site Tiling Analysis. The approach simultaneously discovers polymorphisms and provides quantitative genotype data at 10,000s of loci. It is highly accurate and free from ascertainment bias. We apply the approach to uncover genomic differentiation in the purple sea urchin.

???displayArticle.pubmedLink??? 20403197
???displayArticle.pmcLink??? PMC2884547
???displayArticle.link??? Genome Biol

Genes referenced: fst LOC100887844 pklr

???attribute.lit??? ???displayArticles.show???

	Figure 1. Restriction site tiling analysis identifies polymorphisms and genotypes individuals by hybridization to a custom microarray. Fifty base pair tiles (white circles) are designed to be centered on restriction enzyme cut sites. DNA from an individual is extracted and randomly sheared by sonication. The sample is then divided in half: one part is treated with the restriction enzyme and labeled with green fluorescent dye (Cy3), the other part is treated as a control (without restriction enzyme) and labeled with red fluorescent dye (Cy5). The two parts are mixed and hybridized to the array. This DNA processing and hybridization result in different fluorescent signals reflecting the three possible genotypes for a polymorphic locus: when an individual is homozygous for the cut site (blue triangle) the digested DNA is cut and does not hybridize to the tile, resulting in a high red-to-green ratio (log2 Cy5/Cy3, left panel); however, if an individual is homozygous for a mutation in the cut site (yellow star) then the DNA remains intact and hybridizes to the tile, resulting in high green signal intensity or a low red-to-green ratio (right panel). Heterozygous individuals yield an intermediate red-to-green ratio. Polymorphic loci are identified based on the bi- or trimodal distribution of log ratios across sampled individuals. Individuals can be genotyped based on their log ratio.
	Figure 2. Frequency histograms of signal intensities for experimental and control tiles. (a) Digested DNA (green, labeled with Cy3) and non-digested DNA (red, Cy5) binding to restriction cut site centered tiles. (b) Cy5 signal intensities for negative control tiles (blue, randomly generated tiles that did not match anywhere in the genome according to BLASTN) and positive control tiles (magenta, matching multi-copy ribosomal DNA).
	Figure 3. Polymorphic restriction cut site in pyruvate kinase muscle isozyme across 20 individuals. (a) RSTA array log ratio data separate genotypes of individuals sampled. Cool colored circles represent individuals from Boiler Bay, Oregon; warm colored triangles represent individuals from San Diego, California. The data for each individual are in triplicate. (b) Individual genotypes confirmed by restriction digest gels. Lane 1 is an undigested PCR fragment for size reference, while lanes 2 to 10 are treated with the restriction enzyme; lanes 2, 3, 5, 6, 9, and 10 are from heterozygous individuals; lane 4 is from an individual homozygous for the cut site; lanes 7 and 8 are individuals homozygous for a mutation in the cut site. (c) Genotype data resulting from RSTA can be used to look for differences across populations.
	Figure 4. Principal Components Analysis using RSTA array log ratio data show a signal of population differentiation in a high gene flow species. Symbols represent individuals from Oregon (blue circles) and San Diego (red triangles). (a) All polymorphic coding loci, 6,859; (b) polymorphic coding loci excluding top FST loci, 6,555; and (c) top FST polymorphic coding loci, 304. Patterns were similar for other tiles in non-coding regions.
	Figure 5. Genome-wide distribution of FST values. Open bars show the observed distribution for 12,431 polymorphisms. Solid bars show the mean of 10,000 random permutations. Error bars represent standard deviation for permuted distributions. Numbers in boxes show excess number of loci observed over mean permuted.

References [+] :

Akey, Interrogating a high-density SNP map for signatures of natural selection. 2002, Pubmed