20.02.06 SNP Arrays

Question 1

What is a SNP

Answer 1

• Single nucleotide polymorphism

- A DNA variation occurring commonly within a population (>1%) where there is a single nucleotide change.

Question 2

Is the distribution of SNPs homogenous

Answer 2

No. SNPs occur more frequently in non-coding regions

Question 3

What can be used as a predictor of SNP density

Answer 3

• Microsatellites.

- AT microsatellites are found in regions of significantly reduced SNP density

Question 4

What two types of SNPs are there in coding regions

Answer 4

• Synonymous (do not affect protein sequence, can still affect function)
• Non-synonymous (change amino acid sequence)

Question 5

What two types of non-synonymous changes are there

Answer 5

• Missense

- Nonsense

Question 6

What is gnomAD

Answer 6

• Genome Aggregation Database
• 123,136 exomes
• 15,496 genomes.
• From unrelated individuals

Question 7

3 main components of a SNP array

Answer 7

• Immobilised allele-specific oligonucleotide probes
• Fragmented nucleic acid sequences of target labelled with fluorescent dye
• Detection system (converts probe signal intensity to genotype)

Question 8

What does SNP array signal intensity depend on

Answer 8

• Amount of target DNA in the sample

- Affinity between target DNA and probe (SNP mismatch will reduce binding efficiency)

Question 9

What are the two leading technologies

Answer 9

Illumina and Affymetrix

Question 10

How does the Illumina SNP array work

Answer 10

• Illumina beads have a 50nucleotide sequence attached, which is complementary to the sequence adjacent to the SNP site
• Single base extension that is complementary to the allele carried by the DNA, results in an appropriately coloured signal

Question 11

How does the Affymetrix SNP array work

Answer 11

• 25 nucleotide probes, for both allele
• Location of SNP varies from probe to probe
• Target DNA binds to both probes regardless of allele present
• When the target sequence is complementary to all 25 nucleotides of the probe, the signal strength is strong
• Partial homology will produce a weaker signal

Question 12

How are SNPs chosen for the assay

Answer 12

-Variability in a population

Question 13

Theory of B allele chart

Answer 13

• BB homozygotes have data value of 1.
• AA homozygotes have data value of 0
• AB hets are 0.5
• SNP data is plotted as the B allele frequency
• B allele frequency for each SNP is calculated by B/A+B
  e. g.
• B hom= 2/(0+2)=1.0
• A hom= 0/(2+0)=0
• Hets= 1/(1+1)= 0.5

Question 14

Using B allele chart. How is a duplication shown

Answer 14

• B duplicated= B/(A+B) = 2/(1+2)= 0.666

- A duplicated= 1/(1+2)= 0.333

Question 15

Using B allele chart. How is a deletion shown

Answer 15

• B deleted= B/(A+B) = 0/(1+0)= 0.0

- A deleted= 1/(0+1)= 1.0

Question 16

Using B allele chart. How is a mosaic deletion shown

Answer 16

• 20% B deleted= B/(A+B) = 0.8/(1+0.8)= 0.444

- 20% A deleted= 1/(0.8+1)= 0.555

Question 17

Factors affecting LogR Ratio

Answer 17

• Amplification (larger LogR ratio), amplification of 1 homologoue will result in diluting out of AB alleles.
• MCC. Greater the MCC the greater the divergence away from expected 1.0, 0.5, 0.0 values
• Nullisomy= homozygous loss, leads to a drop in LogR Ratio
• homozygous UPD. Both identical copies of chromosome are inherited from one parent. Every SNP will be hom.
• Copy number neutral LOH. No gains or losses but significant loss of heterozygous SNPs. Indicator of consanguinity or LOH in malignancy

Question 18

Postnatal applications of SNP arrays

Answer 18

• First line test for dev del and/or dysmorphism patients.
• long stretches of homozygosity can unmask potential recessive diseases where clinician suspects a particular gene. Although further testing is needed to confirm the hom variant.

Question 19

Considerations for SNP arrays in prenatal testing

Answer 19

-Do not require sex-matched controls therefore sex of fetus is not needed before set up.

Question 20

SNP arrays in oncology

Answer 20

• LOH and allelic imbalance: to differentiate between cancer subtypes
• CNV: copy gain of 8q, where TPD52 is located (overexpressed in prostate cancer)
• LOH and CNV analysis: looks for UPD. e.g. in acute meyloid leukemia, 20% have UPD.
• Methylation: compare presence and absence of DNA sequences between the methylation sensitive and non-sensitive enzymes.
• allele specific gene expression: use cDNA instead.

Question 21

Limitations of SNP arrays

Answer 21

• Can’t detect balanced rearrangements, gene fusions, whole genome ploidy changes
• Mosaicism under 20-30% is not reliable. SO can’t be used for minimal residual disease detection