Arrays

Question 1

Types of array and applications

Answer 1

1. Comparative genomic hybridisation: CNV
2. Gene expression array
3. Chromatin immunoprecipitation on chip
4. SNP array (snp genotype and CNV)
5. Exon array (detect different forms of splicing)
6. Tiling array (overlapping probes densely represent region)
7. Fusion gene array (cancer fusion transcripts)

Question 2

What is aCGH?

Answer 2

Competitive genomic hybridisation (aCGH) is used to detect copy number changes across the whole genome.

Differential fluorescent labelling of patient and reference DNA and competitive hybridisation (1:1 ratio) to thousands of oligonucleotides immobilised on a slide

A microarray scanner measures the signal intensity of red and green fluorescence for each oligonucleotide probe

Signal intensity is compared after normalisation

Software is used to calculate the logarithm of the ratio of signal (Log2), which are plotted by each chromosome

Circular binary segmentation is then used to detect imbalances

Question 3

Clinical applications of arrays

Answer 3

Constitutional cytogenetics

Prenatal diagnosis

Solid tumours

Products of conception

Malignancies

Question 4

Common cause of recurrent CNVs

Answer 4

Presence of segmental duplications

Non-allelic homologous recombination between low copy repeat sequences flanking the region

Question 5

Resolution of karyotyping

Answer 5

> 5Mb

Question 6

Advantages of aCGH

Answer 6

Rapid and accurate detection of CNV over whole genome

High resolution (min 180Kb in Manc)

Enables deletion/duplication breakpoint mapping and sequencing if necessary

Enabled the discovery of many new syndromes not detectable by karyotyping

Identifies gene content of pathogenic imbalances

Can detect unbalanced products of balanced parental rearragements

Question 7

Limitations of aCGH

Answer 7

Cannot detect balanced rearrangements (reciprocal translocations, inversions)

Does not provide positional information e.g. duplication tandem or insertion? (need FISH)

Two changes on one chromosome (are they on same homologue? implications for recurrence risk)

Detects numerous VUS (interpretation complex)

Cannot detect the order or orientation of rearranged segment (FISH studies required)

May not detect low levels of mosaicism (abnormal cells

Question 8

What is a CNV?

Answer 8

A segment of DNA that differs in copy number in comparison to the reference genome

12% human genome is CNV

Contribute to 7% of genome variability within humans

Question 9

Considerations for clinical interpretation of CNVs

Answer 9

1. Comparison with internal and external databases (known normal variation or known pathogenic/ frequency in normal pop)
2. Association with known syndromes (Unique)
3. Genomic content (genes associated with phenotype? Dosage sensitivity? Does this fit?)
4. CNV size (large CNVs may be benign and small pathogenic!)
5. Follow up studies (inherited from phenotypically normal parent?/de novo? cases of non-penetrance)

Question 10

Follow up studies for abnormal CNV

Answer 10

Confirmation not required

Parental bloods are not routinely investigated if UV identified in proband

If pathogenic CNV detected, parental bloods studied to assess recurrence risk and establish inheritance

Use of FISH is preferable (positional information, may detect balanced rearrangement)

Question 11

Postnatal referral acceptance criteria

Answer 11

Patients showing:

• Developmental delay
• Learning difficulties
• Behavioural disorders
• Dysmorphism
• Multiple congenital abnormalities

Parental bloods after detection of abnormality in proband

Question 12

Prenatal referral acceptance criteria

Answer 12

At least one structural abnormality on scan

Nuchal measurement equal to or greater than 3.5mm in first trimester (or 6 in the second)

IUGR without placental insifficiency

Fetuses with a sex chromosome aneuploidy that is unlikely to explain the ultrasound abnormality

(in Manc arrays are performed if initial QF-PCR is normal and not performed for increased risk of Down’s)

Question 13

Classifying variants as benign (class 1)

Answer 13

Evidence in curated databases of normal variants at a level of >1% in control populations in a least 1 publication (be critical of pop no.) with same gene content, ideally 100% coverage

Comparison with internal database. >3% preferable with 100% coverage. Manchester at least 10 patients with same CNV.

Decipher consensus frequency >1%

Question 14

Likely benign (class 2)

Answer 14

No gene within region of CNV

Evidence at low levels in curated databases of normal variants in control populations

Intronic where no evidence of pathogenicity of same change in lit/databases

min/max co-ordinates more than 20bp from intron/exon boundary

Question 15

VUS (class 3)

Answer 15

Contains genes where there is no published evidence on dosage sensitivity

No clear evidence in literature to support/refute pathogenicity or conflicting evidence

CNV includes gene associated with recessive disorder

Question 16

Likely pathogenic (class 4)

Answer 16

There is limited or emerging evidence published in peer-reviewed high impact journals indicating possible/likely pathogenicity for gene(s) in CNV

Single case reports should not be used to infer likely pathogenicity

Genes within region have function relevant to reason for referral

Large CNVs with phenotype similar to that previously described (unique)

Question 17

Pathogenic (class 5)

Answer 17

CNV contains known disease causing genes with a mechanism of action consistent with CNV change

Known recurrent pathogenic change/known syndrome

CNV or region within it reported as clinically significant in multiple publications

Classification includes clinically significant incidental findings

Question 18

CNVs associated with neurodevelopmental disorders with reduced penetrance

Answer 18

Can be inherited from asymptomatic parent

Outcome to future pregnancies difficult to predict making prenatal diagnosis questionable

Includes loss of:
1q21.1
16p11.2
16p13.11
16p12.1
17q12
15q13.2

Question 19

Incidental findings include

Answer 19

Late onset conditions

Cancer-susceptibility genes