Optimising array design and content for the modern cytogenetic research lab
Ephrem Chin, Duarte Molha, Douglas Hurd, David Cook, Anthony Allen, John Anson
Presented at the American Cytogenetics Conference at Sunriver Resort, OR, USA in June 2016, this poster outlines how the design for the CytoSure™ Constitutional v3 array was developed and optimised to provide CNV detection with exon-level resolution in developmental delay research.
Want to download this as a PDF? Download now or preview the poster below.
Array Comparative Genomic Hybridization (aCGH) has been replacing karyotyping as the method of choice to detect chromosomal aberrations such as gains and losses in developmental delay research. There are a limited number of probes (oligos) that can be attached to an array and therefore it is important that these probes are in the genomic location most likely to detect an important aberration.
The traditional designs, such as OGT’s CytoSure ISCA v2, consisted of probes evenly spaced throughout the genome (the backbone probe set) with increased density of probes for some genes which are regarded as potentially of interest to cytogenetics researchers.
These new designs have taken an evidence-based approach to gene targeting to enable single-exon-level CNV detection in disease relevant genes while retaining backbone coverage.
The ClinGen organisation (formally ISCA, ICCG) has been compiling lists of pathogenic genes and genomic regions accumulated from laboratories around the world1 . The Deciphering Developmental Disorders (DDD) project has added to this knowledge through the study of over 10,000 trio samples using high density arrays and whole exome sequencing (WES)2,3. Both organisations have categorized genes and regions according to the likely pathogenic effect if there is a chromosomal aberration encompassing or within them.
Table 1: DDD and ClinGen gene classification criteria. DD - Developmental Disorders, IF - Intellectual Functioning.
Using both these sets of data we have designed arrays with increased probe density in regions within the categories of highest importance. In some genes the number of probes is such that the level of resolution is predicted to be able to detect whole exon gains or losses. There is significant overlap between the two data sets.
Table 2: Level of coverage of the genes targeted across all three CytoSure Constitutional v3 designs.
An additional 67 decipher syndromic regions and 41 additional ClinGen regions were targeted at high resolution due to their disease relevance. For all other regions not included in the previous categories, a three-tiered backbone design based on the combination of haploinsufficiency (HI) score and number of ClinGen pathogenic CNV in 1Mb regions across the genome. The HI (lod) score and ClinGen pathogenic CNV score was combined into an overall score using a ranks based approach. Combined score was separated into three discrete bins (ranks), low, medium and high to receive differential probe coverage.
Table 3: Coverage of the backbone and genomic region of each of the three CytoSure Constitutional v3 designs.
Probe optimisation is critical to enable exon-level coverage of targeted genes. It is important to eliminate cross-hybridisation to secondary targets and non-specific binding as well as minimizing GC content. We have developed probe design algorithms, and use comprehensive in silico and empirical testing to create the Oligome™ database containing ~26.5 million probes that have been screened for homologies and secondary structures and performance ranked before being included in the database. The Oligome is constantly re-evaluated to enable the development of high-quality array designs, ensuring delivery of the best possible CNV and LOH detection calls. These probes were used to create the CytoSure Constitutional v3 designs, the number of probes used for each gene category are shown in Figure 1.
Figure 1. The number of probes used to target genes from each gene category for each of the three CytoSure Constitutional v3 array designs.
The new array designs are able to detect very small CNV at the exon level even in the 8x60k format (containing the lowest number of probes per array). Shown here are a single exon <500bp duplication (Figure 2), a 2.2kb deletion of 2 exons (Figure 3) covering the aberration, and a 269bp single-exon deletion (Figure 4)*.
Figure 2: A single exon duplication within the MID1 gene detected using CytoSure Constitutional v3 8x60k, minimum size <500bp.
Figure 3: CREBBP (Rubenstein-Taybi) deletion (16p22.2) detected on CytoSure Constitutional v3 8x60k array. The deletion is 2.2kb covering 2 exons.
Figure 4: MED13L single-exon deletion of 269bp detected with CytoSure Constitutional v3 8x60k array.
By giving careful consideration to the targeted gene content it is possible to produce an array design for high-throughput applications which enables the detection of single- and multipleexon micro-deletions and micro-duplications as well as large structural changes. Optimizing probe design and coverage and focusing probes to individual exons combined with using the most up to date information from the most relevant databases and projects is required to achieve this.
Until NGS techniques progress to be able to reliably detect CNV, especially in the 1-3 exon size range, microarrays, using the latest content to guide their design, will remain the best tool for detection of both large and small genomic CNV.
- The Deciphering Development Disorder Study. Large-scale discovery of novel genetic causes of developmental disorders. Nature (2015) 519 p223-228.
- Wright et al. Genetic diagnosis of developmental disorders in the DDD study: a scalable analysis of genome-wide research data. The Lancet (2015) 385 p1305-1314.
* Data courtesy of Dominic McMullan, West Midlands Regional Genetics Laboratory, Birmingham, UK.
CytoSure products are for research use only; not for use in diagnostic procedures.
Request a technical consultation
Do you have a question about what you've just read? Contact us today and one of our technical specialists will be happy to answer any questions you may have.
Download a PDF of this literature.