Arrays Flashcards
What is a DNA microarray?
- Gene chip, DNA chip or biochip array is a collection of DNA probes attached to a solid surface.
- Allows to simultaneously examine copy number changes within a DNA sample.
- Probe-target hybridisation is usually detected and quantified by detection of fluorochrome labelled targets.
What is the most common type of array technique used in diagnostics? What is the principle behind the technique?
- Array CGH is the most common type of array technology in clinical practice.
- It relies on the concept of competitive hybridisation of differently labelled reference and test DNA.
1) . DNA extraction and digestion
2) . Test DNA labelled red with Cy5, ref DNA labelled green with Cy3.
3) . Test and ref DNA combined in 1:1 ratio and left to hybridise to the DNA spotted chip.
4) . Where there is a lack of a DNA region in the test DNA only the ref DNA will bind and it will appear green on scanning. Where there is a gain more test than ref DNA will bind and it will appear red. Yellow indicates no difference between the test and ref DNA (i.e. no losses or gains) in that location.
Very simple, how does a microarray scanner work and what is done with the data?
- It measures the intensity of fluorescence at each of the oligonucleotide spots simultaneously.
- Software allows normalisation, visualisation, breakpoint analysis and comparative analysis of the data. This allows for the accurate detection of CNVs.
What different types of arrays are there?
- Comparative genomic hybridisation: assessing genome content in different cells or closely related organisms.
- Gene expression arrays: assessing gene activity
- Chromatin immunoprecipitation on Chip: combines chromatin immunoprecipitation (ChiP) with DNA microarray.
- SNP array: Identifying SNPs
- Exon array: designed to detect different spicing isoforms
- Tiling array: consits of overlapping probes designed to densely represent a genomic area
- Fusion gene microarray: detects fusion transcripts
What are the basic types of arrays used for assessing copy number variants in cytogenetics?
- BAC arrays: amplified BAC DNA spotted onto glass slides, effectively the same as multiple FISH probes. Resolution about 0.5 to 1 Mb.
- Targeted arrays are designed for specific regions of the genome for the purpose of evaluating targeted segments.
- Oligo arrays: Custom oligos spotted onto glass slides. Higher resolution. Resolution for cytogenetics is about 100-200kb. With increased coverage the resolution is improved.
- SNP array: Detects SNPs. Allele specific oligonucleotide probes are used to detect polymorphisms.
- At present platforms are combining both aCGH and SNP genotyping to optimise diagnostics.
Describe BAC arrays.
BAC arrays: amplified BAC DNA spotted onto glass slides, effectively the same as multiple FISH probes. Resolution about 0.5 to 1 Mb.
Describe targeted arrays.
Targeted arrays are designed for specific regions of the genome for the purpose of evaluating targeted segments.
Describe oligo arrays.
Oligo arrays. Custom oligos spotted onto glass slides. Higher resolution. Resolution for cytogenetics is about 100-200kb. With increased coverage the resolution is improved.
Describe SNP arrays.
SNP array: Detects SNPs. Allele specific oligonucleotide probes are used to detect polymorphisms.
In what areas of cytogenetis can array technologies be applied?
- Constitutional cytogenetics
- Prenatal diagnosis
- Solid tumours
- Products of conception
- Malignancies
Briefly describe the history of arrays.
- In 2004 across the human genome genomic imbalances were defined by several studies. Approximately half of these regions were found to overlap with clones and coincide with segmental duplications. These phenotypic variations show a more dynamic human genome structure.
- The 1st gen CNV map of the human genome was a study of 270 individuals from 4 populations with ancestry from Europe, Africa and Asia. This is the HapMap collection. DNA from these individuals were screened for CNVs using 2 complimentary technologies, SNP based arrays and clone based CGH. A total of 1,447 copy number variable regions (12% of the genome) were defined in these populations. These CNVs contained hundreds of genes, disease loci, functional elements and segmental duplications. Notably the CNVs encompassed more nucleotides content per genome than SNPs thus underscoring the importance of CNVs in general diversity and evolution. The utility of CNVs for genetic disease studies was also demonstrated.
- Another study found that routine karyotype analysis is not sensitive enough to detect subtle chromosome rearrangements less than 5Mbs. The presence of subtle DNA copy number changes was investigated by aCGH in 50 patients with LD and dysmorphism using a DNA microarray constructed from large insert clones spaced at approximately 1Mb intervals across the genome. 12 copy number variations were defined in 12 patients - 7 deletions (6 de novo and 1 inherited from phenotypically normal parent) and 5 duplications (1 de novo and 4 inherited from a phenotypically normal parent). These CNVs ranged in size from those involved in a single clone to regions as large as 14Mb. On the basis of these results it was anticipated that aCGH would become a routine method of genome wide screening for unbalanced rearrangements in children with learning disabilities.
What kinds of things can be performed using aCGH?
- Rapid and accurate screening for subtle DNA copy number changes along the whole genome which cannot be detected by karyotype.
- Precise delineation of deleted or duplicated segments in order to investigate the link between the presence of one unbalanced rearrangement and the observed phenotype.
- Deletion/duplication breakpoint mapping and sequencing. A new way to investigate if some rearrangements could be driven by specific DNA sequences/motifs/conformations.
List some of the referral criteria for aCGH testing.
- Prenatal onset abnormal growth pattern
- Global developmental delay and learning difficulties
- Behavioural problems
- One or more congenital malformations
- Cranio-facial dysmorphism
- Abnormal dermatoglyphic or trichoglyphic patterns
- Family history of multiple miscarriages, learning difficulties, or congenital malformations.
Describe how and why aCGH has replaced karyotyping.
- aCGH at least doubles the detection of chromosome imbalances compared to karyotype in patients with neurodevelopmental disorders.
- Identifies both the location and gene content of pathogenic imbalances.
- aCGH investigations of IMR/Dysmorphic patients is revealing new deletion and duplication syndromes.
- Arrays have also revealed unexpected levels of complexity and imbalance in cytogenetically balanced rearrangements. Approximately 40% of patients with apparently balanced de novo rearrangements and abnormal clinical phenotypes will have imbalances detected by arrays.
- aCGH will detect imbalances which are the unbalanced products of balanced parental rearrangements.
What are the limitations of aCGH?
- Cannot detect ‘balanced’ chromosome rearrangements such as reciprocal translocations and inversions. Particularly the order and orientation of the rearranged segments involved cannot be determined.
- May not detect mosaicism where the proportion of abnormal cells is less than 30%.
Describe Copy Number Variations (CNVs).
- A CNV is a segment of DNA that differs in copy number compared with a reference genome.
- CNVs are found in 12% of the human genome and up to 40% overlap with known genes.
- Contribute up to 7% of genome variability within humans.
- Copy number variation has also been associated with autistic spectrum disorders and idiopathic learning disability.
- Some CNVs have been associated with susceptibility or resistance to disease.
- Some CNVs are common whilst others are rare.
- Many CNVs are benign. Over 29,000 CNVs are catalogued in the database of Genomic Variants.
- Some CNVs are known to be pathogenic.
- Array interpretation has to distinguish innocuous variation from that associated with disease.
What % of the human genome are CNVs found in?
CNVs are found in 12% of the human genome and up to 40% overlap with known genes.
What % of human genome variability is due to CNVs?
Contribute up to 7% of genome variability within humans.
Describe the considerations for clinical interpretation of CNVs.
1) . Comparison of CNV with internal and external databases.
2) . Association of CNV with well established syndromes.
3) . Genomic content.
4) . CNV size.
5) . Follow up studies.
6) . Inheritance of CNV.
How would you go about the comparison of a CNV with internal and external databases when investigating the clinical significance of a CNV?
- Look at the frequency of the CNV in internal database.
- Comparison with the Database of Genomic Variants (DGV).
- Is the CNV associated with known syndromes (see Decipher, PubMed, OMIM, ISCA, ECARUCA).
List some useful resources for array interpretation.
- UK Best Practice Guidelines - very general and their usefulness is limited.
- American College of Medical Genetics gives more detailed standards and guidelines for interpretation and reporting of postnatal constitutional CNVs.
- Another useful resource is ‘Practical guidelines for interpreting copy number gains detected by high resolution array in routine diagnostics’. This was a paper published in European Journal of Human Genetics in 2012.
- Also useful is ‘Diagnostic Interpretation of Array Data Using Public Databases and Internet Sources’ published in Human Mutation.
What considerations should be applied when assessing the genomic content of CNVs to aid interpretation of clinical significance?
Consideration of Genomic content in CNV:
- Are there genes relevant to the phenotype?
- Are the genes dosage sensitive? Review Decipher, OMIM and relevant publications.
- If there is a loss, is a phenotype associated with haploinsufficiency? This may not have an effect when their is gain.
- CNVs involving genes with dominant mutations may not have relevance or result in a different phenotype.
- Single copy deletions of a recessive gene may only suggest carrier status, depending on a mutation on the other homologue.
How important is CNV size? Explain your answer.
CNV size is not always important:
- Not all large CNVs are pathogenic!
- There are even large euchromatin chromosomal variants that are known to be benign.
- Very small CNVs can be pathogenic.
- The threshold for detecting imbalances should be based on array performance balancing specificity and sensitivity.
- Potential pitfalls of using arbitrary size cut-offs.
- Careful consideration of gene content of unbalanced regions is required when interpreting aCGH data.
Describe what follow up studies may be conducted after aCGH.
Follow up studies for an abnormal CNV:
- In the earlier stages of the aCGH service, all reportable CNVs were confirmed in proband either with FISH or with MLPA/PCR to assess artefacts.
- Confirmation is not required in high coverage arrays and in platforms which combine aCGH and SNP arrays.
- Around 25% of referrals have imbalances which would require parental bloods to assist interpretation (unknown significance CNVs).
- 15-20% of there imbalances turn out to be de novo and depending on gene content may be causative.
- Parental bloods are tested to ascertain their recurrence risk (parents may carry a balanced rearrangement or can be asymptomatic in cases of pathogenic CNVs with reduced penetrance).
- FISH is preferable - e.g. the position of signals on chromosomes in a balanced parent can unravel submicroscopic balanced rearrangements (translocations, inversions, insertions) which would confer high recurrence risk.
Are familial CNVs clinically relevant?
- Although in clinical practice it is considered that if a CNV is inherited from a phenotypically normal parent then it is likely to be benign, this is not always correct however. This depends on the penetrance of the CNV.
- Syndromic CNVs have reduced penetrance and variable expressivity.
- De novo CNVs are more likely to be causal but this is by no means always the case.
- Inherited copy number gains could also be causal as may often be inherited from an asymptomatic parent.
- Phenotype may depend on the co-occurrence of other CNVs and mutations.
What are the essential steps in classifying a CNV as ‘benign’?
1) . Comparison with the Database of Genomic Variants (DGV).
2) . >1% frequency and 100% coverage with same gene content and similar dosage (gain or loss) in at least one publication - happy it is benign and no need to report it.
3) . Sex for X-linked CNVs.
4) . Comparison with Internal Database. >3% and preferable 100% coverage. May represent a local polymorphism or be platform related.
How do we decide that a CNV is clearly pathogenic?
- Clinical significance is established in peer reviewed publications.
- Includes large CNV which overlap a smaller interval with clearly established significance.
- Smaller CNVs which include critical causative genes of known clinical syndromes.
Give some examples of pathogenic CNVs.
All known cytogenetic syndromes: DiGeorge, Prader Willi, Williams, Turner, Potocki-Shaffer and Kleefstra syndromes, MECP2 duplications, ring chromosomes and imbalances derived from balanced rearrangements. Some examples:
- A de novo loss of 3.75Mb as 22q13.31-q13.33 which overlaps the 22q13.3 syndrome (patients show some degree of mental retardation, and absence of expressive speech).
- De novo amplification of a 0.79Mb at Xp11.22. This includes HUWE1 gene which is linked to non syndromic X-linked learning difficulties.
- Maternally inherited loss of 750kb of 22q11 within the DiGeorge critical region.
What phenotype will a de novo loss of 3.75Mb as 22q13.31-q13.33 which overlaps the 22q13.3 syndrome region produce?
A de novo loss of 3.75Mb as 22q13.31-q13.33 which overlaps the 22q13.3 syndrome (patients show some degree of mental retardation, and absence of expressive speech).
What phenotype will de novo amplification of a 0.79Mb at Xp11.22 produce?
- De novo amplification of a 0.79Mb at Xp11.22. This includes HUWE1 gene which is linked to non syndromic X-linked learning difficulties.
Give some examples of CNVs associated with neurodevelopmental disorders of reduced penetrance.
These include: - Loss of 1q21.1 (proximal and distal) - Loss of 16p11.2 (proximal and distal) - Loss of 16p13.11 - Loss of 16p12.1 - Loss of 17q12 - Loss of 15q13.2 Gains in some of these regions are also associated with milder and different phenotypes.
There are also emerging syndromes such as the 2q11.1-2 deletion and 15q11.2 deletion which has only 10% penetrance and is not universally accepted as syndromic.
The presentation of these syndromes is usually variable and can be inherited from an asymptomatic parent. Inheritance studies would establish the recurrence risk for parents.
The outcome of future pregnancies which carry these CNVs is difficult to predict which makes prenatal diagnosis questionable.
What is the phenotype related to loss of 1q21.1 (proximal and distal)?
Neurodevelopmental disorder of reduced penetrance.
