Copy Number Detection (Cytogenetics)

Question 1

What are the main cytogenetics methods for detecting copy number in DNA samples?

Answer 1

1. G-banding
2. FISH
3. BAC-arrays
4. Oligo-arrays
5. SNP-arrays

Question 2

Briefly describe how G-banding is used for copy number detection.

Answer 2

Manipulation of cell cycle to obtain metaphase cells enabling comparison of banding pattern between homologues.

Question 3

Advantage of using G-banding for copy number detection.

Answer 3

1. Whole genome screen
2. Detects balanced/unbalanced rearrangements
3. Provides positional information
4. Detects mosaicism
5. Relatively robust and inexpensive
6. Determine structural
7. rearrangements e.g. ring chr 20

Question 4

Disadvantage of using G-banding for copy number detection.

Answer 4

1. Low resolution (>5Mb)
2. Labour intensive
3. Slow turnaround time
4. Unable to detect UPD
5. Requires dividing cells and manipulation of the cell cycle
6. Risk of cultural artefacts e.g. prenatal cases
7. Some abnormalities (usually mosaic aneuploidies) not detected in cultured cells

Question 5

Examples of using G-banding for copy number detection.

Answer 5

1. To detect pre/postnatal aneuploidy
2. To detect unbalanced rearrangements in dysmorphic child
3. To detect balanced rearrangements in recurrent miscarriage couple
4. Detection of abnormal clones in cancer
5. Detect clinically significant structural rearrangementse.g. t(15;17) in APL

Question 6

Briefly describe how FISH is used for copy number detection.

Answer 6

• Fluorescently labelled ssDNA probes are hybridised to specific denatured DNA sequences e.g. metaphase spreads or interphase nuclei.
• Probes typically target specific loci e.g. centromere, telomere, locus-specific, whole chromosome paint or BACs

Question 7

Advantage of using FISH for copy number detection.

Answer 7

1. Positional information if metaphases analysed
2. Detects mosaicism
3. Detects ploidy
4. Aids interpretation of G-banding
5. Fast turnaround time (able to perform shorter hybridisation in urgent cases e.g. ?T18)
6. Higher resolution than G-banding
7. Large number of individual cells can be examined
8. Can be used to analyse single cells
9. Probes available for almost any genomic region

Question 8

Disadvantage of using FISH for copy number detection.

Answer 8

• Targeted test
• Probes can be expensive
• Cannot detect MCC in certain cases
• Cannot detect UPD
• Limited number of probes can be used at one time, only 2 or 3 colours possible
• Interphase FISH provides no positional info
• May require metaphases
• Atypical rearrangements may be normal by FISH e.g. some t(15;17) arrangements
• Microdupsmay be undetected due to limited resolution on metaphase spread
• Co-localisation can occur: two signals overlap and appear as one
• Cross-hybridisation can occur: probe binds to regions with repetitive sequences

Question 9

Examples of using FISH for copy number detection.

Answer 9

• Aneuploidy screen
• Microdeletion/duplication detection
• Aid G-banding e.g.origin of marker chr
• Detection of cryptic rearrangements e.g.t(12;21) in cancer
• Used if G-banding fails e.g. AML screen using disease specific probes
• Detection of mosaicism
• Detection of abnormal clones/clonal evolution in cancer
• Gene amplification e.g. HER2 in br. Ca. or N-MYC in neuroblastoma
• Gene deletions e.g. TP53 or ATM
• PGD
• Variant gene fusionse.g. BCR-ABL1 in CML
• Post-transplant chimaerism monitoring

Question 10

What is Array Comparative Genomic Hybridisation (arrCGH) ?

Answer 10

Array Comparative Genomic Hybridisation (CGH) is a molecular technique based upon c_ompetitive hybridisation_ of a test sample and a control sample to identify any possible gains or losses at particular sites of interest.

Within a genetic context, arrays can be used on either DNA or RNA samples

Question 11

Give an overview of the array CGH protocol

Answer 11

1. Patient and control DNA is labelled with fluorescent dyes and applied to the microarray (Cy3 vs Cy5)
2. Labelled DNA is cleaned, combined and precipitated.
3. Pre-hybridisation to Cot-1 DNA: hybridises to repetitive sequences in genome and prevents cross-hybridisation when on the array platform.
4. Sample then hybed to array: Patient and control DNA compete to hybridise to the array probes. Automated and highly controlled in incubator or oven (rotates slide for even hyb)
5. Dual wavelength laser scanner is used to measure fluorescent signals on array
6. Software analyses the data and generates plots.

Question 12

What are the expected Cy3 vs Cy5 ratios for various copy number changes detected by arrCGH?

Answer 12

• Equal hybridisation gives 1:1 ratio of red to green (the colours read from the fluorescence; after normalising for variation already mentioned).
• Deletion gives excess of control DNA sample colour.
• Duplication gives excess of patient DNA sample colour

Question 13

How are the Cy3 vs Cy5 ratios interpreted to gain information about copy number value?

Answer 13

• In ratio world the raw values are 2.0 for a doubling of the test sample and 0.5 for half the test sample
  
  • i.e. reduction ratios are squished between 1 and 0 whereas duplication ratios will be plotted further and further away from 1.
  • This means that the plot is asymmetrical and exaggerates the difference between equal magnitude gains and losses
• Log transformation turns the equal hybridisation (ratio 1.0) value to 0.
• Means that the plot of deletions and duplications becomes symmetrical around 0.
• Doubling of test DNA = log(2)2 = 1
• Halving of test DNA = log(2)0.5 = -1

Question 14

What are the log 2 values for heterozygous deletions and duplications detected by arrCGH?

Why are duplications ‘harder’ to detect by arrayCGH than deletions?

Answer 14

• To do with the log2 values generated for dels vs dups
• Deletion of test DNA = log(2)0.5 = -1
• Duplication of test DNA = log(2)1.5 = 0.58
• Thus, the magnitude of movement away from normal copy number is not as great for duplications

Question 15

What key quality criteria are assessed when determining the success of an arrCGH experiment?

Answer 15

• Inclusion rate of 95+%, ie. the percentage of DNA spots (features) on the array that show good hybridisation – this is a measure of technical reproducibility
• Standard deviation of autosome of <0.075 – a measure of the variation of replicate features across the array – this is a measure of biological reproducibility

Question 16

Briefly describe how BAC-arrays are used for copy number detection.

Answer 16

• Patient and reference DNA labelled with different fluorescent dyes
• Then hybridised to different BAC clones (on a glass slide) spread throughout the genome

Question 17

What types of probes are used on BAC arrays and how are they produced?

Answer 17

• Large PAC, cosmid and fosmid clones of approx. 150-200kb in size are used (derived from the mapping stages of the human genome project).
• BAC clones are propagated in vectors in bacteria, purified and amplified and then spotted onto a glass slide using ultra fine needles.
• Multiple copies of each BAC are spotted onto the array and distributed across the array.

Question 18

What are the advantages of using the large probes found on BAC arrays?

Answer 18

Due to the large size, BACs are very stable and hybridisation is specific (high signal to noise ratio).

Question 19

What advances did BAC arrays bring compared to the technology used prior to their introduction?

Answer 19

• Whole genome BAC arrays developed which include a backbone clone set spaced at 500kb-1Mb intervals across the genome.
• The greater coverage of whole genome arrays compared to the initial targeted arrays developed increased the abnormality detection rate.

Question 20

Advantage of using BAC-arrays for copy number detection.

Answer 20

• Whole genome screen
• Can detect unbalanced copy no changes to ~1Mb resolution
• Can concentrate probes to particular regions of interest e.g. gene rich (can design custom arrays to target specific areas/genes as well as providing a lower resolution coverage over the whole genome)
• Simultaneously process large numbers of samples
• Cheap and reliable - still widely used for PGS (24sure)

Question 21

Disadvantage of using BAC-arrays for copy number detection.

Answer 21

• No positional/structural information
• Doesn’t detect balanced rearrangements/ ploidy
• Lower resolution than other platforms (0.5-1Mb detection levels)
• Expensive
• Labour intensive
• Difficulty with interpreting some results
• Resolution not even across the genome
• Requires relatively large amounts of good quality DNA

Question 22

Examples of using BAC-arrays for copy number detection.

Answer 22

• Largly superseded by higher res arrays
• Still widely used for PGS - 24 sure array
• Genome screen of individuals with idiopathic mental retardation, dev. del, dysmorphism
• Prenatal samples with abnormal scans/solid tissue samples
• Interpretation of abnormal karyotypes

Question 23

Briefly describe how Oligo-arrays are used for copy number detection.

Answer 23

• Patient and reference DNA labelled with different fluorescent dyes.
• Synthetic oligos used in hybridisation stage which are smaller than BAC clones therefore allowing greater coverage.

Question 24

What are the characteristics of he probes used on oligo arrays?

Answer 24

• Made from synthetic single-stranded oligonucleotide fragments, ranging from 25- 85bp in size.

Question 25

What are the different manufacturing methods for producing oligo arrays?

Answer 25

1. Pre-synthesized oligos _spotted_ on glass slides (usually used to produce custom-made oligo-arrays) 2. Oligos are _synthesized directly onto surface_ (in situ synthesis) using laser-directed photochemistry (Nimblgen/Affymetrix eg Gene Chips) or inkjet technologies (Agilent/OGT) 3. _Bead technology_ where oligos, with a unique sequence of ~ 25bp (address) at the 3’ end, are attached to silica beads. The beads are then randomly deposited into wells on a substrate (i.e glass slide).

Question 26

What advantages do the characterisics of the oligo probes themselves bring over BAC probes?

Answer 26

1. Small size allows them to be packed more densely accross the genome, therefore achieving a far greater number of probes in a given portion of the genome. Enables a far higher resolution to detect copy number change - approx. 50-200Kb 2. Because they can be manufactured in situ they can be placed physically closer to each other on a slide, enabling many more probes per slide = much cheaper per patient e.g. 8x60K array.

Question 27

Advantages of using Oligo-arrays for copy number detection.

Answer 27

* High resolution imbalance detection (~50-200Kb depending on coverage) * Can concentrate probes to particular regions e.g. gene rich (can design custom arrays to target specific areas/genes as well as providing a lower resolution coverage over the whole genome) * Simultaneously process large numbers of samples * Analysis of single cells possible (with whole genome amplification) * Uses DNA from uncultured or cultured material

Question 28

Disadvantages of using oligo-arrays for copy number detection.

Answer 28

* No positional information obtained * Doesn’t detect balanced rearrangements/ploidy * Difficulty with interpreting some results-CNVs of unclear significance * Expensive * Labour intensive * Requires relatively large amounts of good quality DNA * Low-level mosaicism not detected.

Question 29

Examples of using oligo-arrays for copy number detection.

Answer 29

* Genome screen of individuals with idiopathic mental retardation, dev. del, dysmorphism * Interpretation of abnormal karyotypes

Question 30

What are SNPs and what are their characteristics, including genomic density, distribution etc?

Answer 30

Single nucleotide polymorphism (SNP) * A DNA sequence variation occurring commonly within a population (e.g. \>1%) in which there is a single nucleotide change * Approximately 50 million SNPs have been found in the human genome. * SNPs occur in non-coding regions more frequently than in coding regions * SNP density can be predicted by the presence of microsatellites: long (AT)(n) repeat tracts i.e. reduced CG content, tending to be found in regions of reduced SNP densit SNPs in the coding region are of two types: * Synonymous * Nonsynonymous

Question 31

Briefly describe how SNP-arrays are used for copy number detection.

Answer 31

* Array-based technology but platform does not require deferential labelling of test and reference samples for comparative hybridisation. * Test sample is hybridised to SNP specific oligos present throughout the genome. * Relative intensity of the A and B-alleles at a SNP loci indicate zygosity/copy number.

Question 32

Describe how the Illumina SNP array platform works.

Answer 32

1. Bead based oligo probe ~50bp in length designed to be immediatley adjacent to a SNP site. 2. A single base extension of a fluorescently labelled nucleotide that is complementary to the allele carried by the test DNA is then incorporated 3. The result is the appropriately coloured signal. 4. Beads are randomly distributed onto a slide so their genomic location has to be 'decoded' during the analysis 5. Beads are very small in size enabelling millions of beads, and thus probes to be distributed onto a single slide giveng very high resolution arrays.

Question 33

Describe how the Affymetrix SNP array platform works.

Answer 33

1. Oligo probes are synthesised in situ on a solid slide 2. The genomic location of the probe and it's coordinates on the array are recorded during manufacturing 3. The SNP site is positioned to be in the centre of the oligo 4. There are probes for the major ad minor allele of each SNP 5. The test DNA will bind with the highest affinity where there is a 100% with the probe. 6. Thus, if hom then most binding will be on the major allele probe with only a small amount of test DNA binding to the mismatched minor allele 7. If het then ~50:50 binding of the major and minor alleles is seen

Question 34

How are the results of SNP arrays interpreted and how does this differ to array CGH?

Answer 34

The relative intensity of the A and B-alleles at a SNP loci indicate zygosity/copy number. * If the intensity of A and B is the same i.e. ratio is 1:1 then the SNP is predicted to be heterozygous * If A = 100 then hom ref allele * If B = 100 then hom alt allele With array CGH it's the relative intensity of a test vs a reference sample that is assessed to determine copy number.

Question 35

What SNP array profile is characteristic of a normal copy number state i.e. diploid?

Answer 35

In a normal diploid individual, you get a balance of AA, BB and AB i.e. a normal allelic balance The heterozygous line (really A-B) is also known as the allelic balance line.

Question 36

What SNP array profile is characteristic of a _heterozygous duplication_?

Answer 36

* In this case there is a heterozygous gain of material we lose the allelic balance line as we never see the A and B alleles in equal proportions * i.e. either have a gain of A or a gain of B = pushed out the tracks * N.B. When we have a gain (i.e. 3 alleles choices , you get 4 lines)

Question 37

What SNP array profile is characteristic of a _heterozygous deletion_?

Answer 37

* In this case there is a heterozygous loss of material. * Here we loose the allelic balance line as either have a loss of A or a loss of B * i.e. cannot have an AB heterozygous line and the tracks are pushed inwards. * N.B. When we have a loss (i.e. 2 allele choices, you get 2 lines)

Question 38

What SNP array profile is characteristic of a copy neutral Loss of Heterozygosity (LOH)?

Answer 38

Terminology, ‘absence of heterozygosity’ important as it does not imply any mechanism, see only AA, or BB patern, nothing in the AB heterozygosity track indicating heterozygosity.

Question 39

Advantages of using SNP-arrays for copy number detection.

Answer 39

* High resolution * Can detect copy number changes and also genotype to reveal LOH and UPD

Question 40

Disadvantages of using SNP-arrays for copy number detection.

Answer 40

* Doesn’t detect balanced rearrangements * No positional/structural information * Coverage dependent on SNP locations * Difficulty with interpreting some results * Expensive * Labour intensive * Can not detect low-level mosaicism

Question 41

Examples of using SNP-arrays for copy number detection.

Answer 41

* Screen for LOH in cancers * Possible application to detect UPD - only uniparental isodisomy. * Could elucidate non-paternity or incidental findings * Whilst the high resolution is not ideal for prenatal investigations SNP arrays do not require sex-matched controls, therefore the sex of the prenatal sample is not needed before set up.

Question 42

What is the real-world utility of detecting LOH in postnatal constitutional applications?

Answer 42

* Although SNP arrays provide LOH data, it is _only relevant_ to report such cases i_f a UPD syndrome is suspected_ in the clinical details (e.g. Prader-Willi syndrome). * It is not usually relevant to report consanguinity. * Any UPD result should be confirmed by a second method e.g. MS-MLPA/PCR or microsats. * Long stretches of Homozygosity may be useful to unmask potential recessive diseases where the clinician suspects a particular gene.

Question 43

Compare the _resolution_ of different array platforms.

Answer 43

* BAC: 0.5-1Mb * Oligo: 50-200kb * SNP: 10kb or less * SNP + Oligo: 50-200kb

Question 44

Compare the key _advantages_ of the different array platforms.

Answer 44

* BAC: * Less CNVs detected of uncertain clinical significance. High signal to noise ratio * Oligo: * Multiple sub arrays, dye swap not required so cheaper per patient * SNP: * Can detect LOH and UPDs as well as CNVs * SNP + Oligo: * Strategic placing of oligos located in areas of segmental duplications. Can detect CNVs more effectively as well as LOH/UPD.

Question 45

Compare the key _disadvantages_ of the different array platforms.

Answer 45

* BAC: * Small abnormalities may be missed. LOH not detected. Expensive per patient. Dye swap required. * Oligo: * More VUSs detected. Lower signal to noise ratio than BACs. * SNP: * Areas of segmental duplication don’t contain SNPs. Less accurate detection of CNVs. Parental samples required for heterodisomy. Lower signal to noise ratio than oligos. Expensive. * SNP + Oligo: * Parental samples required for heterodisomy. Expensive.

Question 46

Compare the _main diagnostic application_ of the different array platforms.

Answer 46

* BAC: * Prenatal testing. PGS (24 sure). POC analysis. * Oligo: * Postnatal (DevDel, IS, ASD). Prenatal (Some labs, EACH project). POC analysis * SNP: * Postnatal as above (higher res). Cancer. PGD (Karyomapping array) * SNP + Oligo: * Postnatal as above (higher res). Cancer.