JMG Online First, published on September 30, 2005 as 10.1136/jmg.2005.037648 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Prenatal detection of unbalanced chromosomal rearrangements by array-CGH

Lisa Rickman1#, Heike Fiegler2, Charles Shaw-Smith1, Richard Nash3, Vincenzo Cirigliano4, GianfrancoVoglino5, Bee Ling Ng2, Carol Scott2, Joanne Whittaker3, Matteo Adinolfi6, Nigel P Carter2, Martin Bobrow1

1 University of Cambridge Department of Medical Genetics, Addenbrooke’s Hospital, Hills Road, Cambridge, CB2 2QQ, UK; 2 The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SA, UK; 3 Regional Genetics Laboratories, Kefford House, Maris Lane, Trumpington, Cambridge, CB2 2FF, UK; 4 Departament de Genètica Molecular, General Lab, 08021, Barcelona, Spain; 5 Molecular Genetics and Cytogenetics Lab, Promea-Day Surgery, 1026, Turin, Italy; 6 The Galton Laboratory, University College London, London, NW1 2HE, UK.

#Corresponding author: Dr Lisa Rickman, The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge, CB10 1SA, UK. Telephone +44 (0)1223 494842 Fax +44 (0)1223 494919

Email [email protected] http://jmg.bmj.com/

on September 25, 2021 by guest. Protected copyright.

1

Copyright Article author (or their employer) 2005. Produced by BMJ Publishing Group Ltd under licence. J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

ABSTRACT Introduction: Karyotype analysis has been the standard method for prenatal cytogenetic diagnosis since the 1970’s. Although highly reliable, the major limitation remains the requirement for cell culture, resulting in a delay of as much as 14 days to obtain test results. Fluorescence in situ hybridisation (FISH) and quantitative fluorescent PCR (QF- PCR) rapidly detect common chromosomal abnormalities but do not provide a genome- wide screen for unexpected imbalances. Microarray based comparative genomic hybridisation (array-CGH) has the potential to combine the speed of DNA analysis with a broad capacity to scan for genomic abnormality. Methods: We have developed a genomic microarray of approximately 600 large-insert clones designed to detect aneuploidy, known microdeletion syndromes and large unbalanced chromosomal rearrangements. This array was tested alongside an array with an approximate resolution of 1Mb in a blind study of 30 cultured prenatal and postnatal samples with microscopically confirmed unbalanced rearrangements. Results: At 1Mb resolution, 22/30 rearrangements were identified, whereas 29/30 aberrations were detected using the custom-designed array, due to the inclusion of specifically chosen clones to give increased resolution at genomic loci clinically implicated in known microdeletion syndromes. Both arrays failed to identify a triploid karyotype. Thirty normal control samples produced no false positive results. Analysis of 30 uncultured prenatal samples showed that array-CGH is capable of detecting aneuploidy in DNA isolated from as little as 1ml of uncultured amniotic fluid; 29/30 samples were correctly diagnosed, the exception being another case of triploidy. Discussion: These studies demonstrate the potential for array-CGH to replace conventional cytogenetics in the great majority of prenatal diagnosis cases.

Keywords: prenatal, array-CGH, aneuploidy, microdeletion

http://jmg.bmj.com/

on September 25, 2021 by guest. Protected copyright.

INTRODUCTION Mainly as a result of screening programmes for the prenatal detection of chromosome abnormalities, approximately 40,000 amniocentesis and chorion villus samples are

2 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from processed annually in the UK 1. The vast majority of these samples (around 90-95%) yield a normal karyotype by full microscopic analysis. A small proportion of cases reveal a chromosome abnormality, about 80% of which are autosomal trisomies for chromosomes 13, 18 and 21. The remaining abnormal karyotypes involve sex chromosome copy number changes and structural chromosomal rearrangements, such as deletions, duplications, inversions and balanced and unbalanced translocations.

Microscopic analysis has been the gold standard for prenatal diagnosis since the development of chromosome banding techniques in the late 1960’s2. Although highly reliable, this procedure has a number of limitations (reviewed in3). Due to the need for cell culture, the average reporting time for results in the UK can be up to 14 days. In addition, microscopic karyotyping is labour-intensive and thus costly, requires skilled interpretation and is not easily amenable to automation. The resolution is limited, with deletions and duplications <10Mb not reliably being detected. Although high resolution methods have been shown to detect abnormalities of 3-5Mb, these procedures are not suitable for routine screening applications (reviewed in4). When a structural chromosome abnormality is suspected, techniques such as fluorescence in situ hybridisation (FISH) or quantitative fluorescent PCR (QF-PCR) can be deployed, but a significant proportion of structural changes are not anticipated at the time of sample collection.

In attempts to overcome some of these limitations, alternative methods for aneuploidy detection based on FISH and QF-PCR have been applied to prenatal diagnosis. QF-PCR utilises primer pairs designed to amplify sequences at several polymorphic loci in a single reaction, and is a rapid, efficient and inexpensive method that is readily amenable to automation5-7. FISH screening for common aneuploidies has also been applied to prenatal testing8. The major limitation of QF-PCR and FISH compared to microscopic karyotype analysis is that they may not detect unbalanced chromosomal rearrangements such as http://jmg.bmj.com/ microdeletions, which although uncommon, account for approximately 1-2% of abnormalities detected by microscopic analysis of prenatal samples, and can have serious clinical consequences9. Both of these techniques have been validated and applied to clinical samples, generally in addition to, rather than replacing, microscopic analysis.

Comparative genomic hybridisation (CGH) was developed as a genome-wide screening on September 25, 2021 by guest. Protected copyright. strategy for detecting DNA copy number imbalances10. The DNA content of a test and reference genome are compared by differentially labelling the genomic DNA with distinct fluorochromes, before competitively hybridising the labelled samples onto normal metaphase chromosomes and analysing the resulting ratio of the fluorochromes. While CGH has been used mainly to analyse the DNA content of tumours as a tool in cancer research11, the technique has recently been shown to be valuable for the detection of copy number imbalances in foetal tissue following loss of pregnancy12. However, CGH still requires metaphase chromosomes as targets for hybridisation, limiting high resolution methods (HR CGH) to around 3Mb13.

Microarray based CGH (array-CGH) is similar in principle to conventional CGH14, 15, but uses arrayed DNA sequences instead of metaphase chromosomes as targets for hybridisation, thus providing a direct link between detected aberrations and the physical

3 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from and genetic maps of the human genome. Array-CGH has a number of significant potential advantages over conventional prenatal testing, providing a technique that is not only sensitive and comprehensive, but may be amenable to automation, thus decreasing cost, labour and the reporting time of results.

Array-CGH has already been shown to be a useful tool in clinical genetics for detecting deletions and duplications in patients with mental retardation/learning difficulties beyond the limits detectable by microscopy16, 17; and for the analysis of individuals with known chromosome abnormalities using a custom-designed microarray18. Array-CGH analysis of foetuses with multiple malformations identified genomic rearrangements which had not been observed by karyotype analysis in around 16% of cases 19. In a study of products-of-conception from spontaneous miscarriages using a low-density array containing targeted clones of clinical significance, array-CGH was able to detect all abnormalities previously identified by microscopic karyotype analysis, and detected additional abnormalities in approximately 10% of cases20. The technique therefore holds some promise of combining the speed, sensitivity and potential for partial automation of a DNA based test, with the genome screening characteristics of microscopic karyotyping.

Although it is becoming accepted that array-CGH will have a place in clinical genetic testing, it is far from clear how this will best be applied. The coverage and resolution of array-CGH are dependent on the design and density of the array used. Although superficially appealing, an array covering the entire genome at very high resolution would have potential disadvantages in clinical use: more array probes are likely to generate a higher number of false positives and large arrays are more expensive to fabricate, quality control and interrogate. Recent investigations showing significant levels of copy number polymorphism in normal populations21, 22 reinforces the desire to only test a limited number of clones, whose results do not give rise to needless complications http://jmg.bmj.com/ in interpretation. We reason that, particularly for prenatal screening when time for further investigation is limited and ambiguous results cause severe anxiety, the ideal array would contain the minimum number of clones that will deliver the required diagnosis.

We have therefore designed and carefully validated an array of approximately 600 large- insert clones, concentrated on areas of known clinical significance with dense on September 25, 2021 by guest. Protected copyright. representation across the common microdeletion regions, and with a lower representation of about one clone per 10Mb over the remainder of the genome to detect unexpected major chromosome imbalance. We have compared the performance of this array with that of a previously described array of 1Mb resolution23, by parallel, blind analysis of 30 pre/postnatal samples known to have unbalanced rearrangements. Because a major advantage of array technology would be lost if cell culture were required, we have further tested the array on DNA extracted from 30 samples of 1-2 ml uncultured amniotic fluid (many with abnormal chromosome constitution), surplus to requirements after diagnostic testing (publication in progress).

We believe these results provide proof of principle that array-CGH utilising a purpose designed array is rapid. Although better than cytogenetics in some respects and worse in others, it will have about the same overall capacity to detect clinically relevant

4 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from chromosome abnormalities as conventional cytogenetics. It has the potential to replace karyotyping for prenatal cytogenetics, but we do not claim the current array is ideal, and further work is required to reach a consensus on the optimum configuration of an array for clinical use. Large trials will be needed to demonstrate sensitivity and specificity in clinical operating conditions.

METHODS Array design and production The prenatal array described in this study was developed using published protocols23. In brief, large-insert BAC and PAC clones were chosen from the published Golden Path of the human sequence, to cover each chromosome at a resolution of one clone every 10Mb. Additional clones were selected for the major common microdeletion syndrome regions, as far as possible covering identified critical regions and microdeletion breakpoints with overlapping clones. Approximate clone locations are shown in fig 1 (clone list in supplementary data). Isolated clone DNA was first amplified by degenerate oligonucleotide-primed PCR (DOP-PCR), followed by secondary PCR with a 5′-amine- modified primer. Array clones were spotted in duplicate onto CodeLinkTM activated glass slides. The 1Mb resolution arrays used in this study are as published previously23.

Array-CGH DNA labelling and array hybridisation DNA samples (0.225µg (chromosome ‘add-in’ experiments, normal controls and cultured samples) or 0.100 µg (uncultured samples) for the prenatal array: 0.45µg for the1Mb resolution array) were labelled and the microarray hybridisation performed as described previously.24

Image acquisition and data analysis http://jmg.bmj.com/ Arrays were scanned using an Agilent scanner (Agilent Technologies, West Lothian, UK) and images quantified using SPOT software25. Further analysis was performed using a custom-designed Excel spreadsheet, in which data were normalised by dividing the ratio of each spotted clone by the median ratio of all autosomal clones. Spots were then excluded from further analysis where duplicate spot ratios differed from the duplicate mean by >10% or where the fluorescent intensities were below those of Drosophila on September 25, 2021 by guest. Protected copyright. control spots.

Array validation Chromosome ‘add-in’ experiments To test the specificity and sensitivity of clones represented on the prenatal array, a series of experiments were performed in which flow-sorted chromosome DNA was added to self-versus-self hybridisations to simulate gain of each chromosome in turn23. Chromosomes were flow-sorted as described26 and DNA isolated from aliquots containing around 250,000 chromosomes using previously published methods27. Prior to labelling, DNA representing either approximately 2 or 4 additional copies of an individual chromosome was combined and labelled together with reference DNA and competitively hybridised against the same reference DNA. Following hybridisation and data analysis, the standard deviation for each array was calculated from the ratios

5 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from comprising the 95th percentile. Thresholds were placed at +/- 4 standard deviations from the median of the 95th percentile values. Using these settings, a clone would be expected to exceed the threshold once in every 24 hybridisations as a result of statistical variation. Clones exceeding the thresholds set for the individual experiment on chromosomes other than the ‘add-in’ chromosome were then identified and excluded from subsequent analyses. This method was used to identify potentially mismapped clones and clones hybridising to multiple genomic loci.

Normal controls DNA isolated from 30 normal, healthy blood donors was hybridised against reference DNA. As described for the series of chromosome 'add-in' experiments, the standard deviation for each array was calculated from the ratios comprising the 95th percentile. Thresholds were placed at +/- 4 standard deviations from the median of the 95th percentile values and clones exceeding the thresholds set for the individual experiment identified.

Patient samples Cultured samples Samples were selected from amongst those available with consent, to represent a broad spectrum of cytogenetic abnormalities including autosomal trisomy, sex chromosome abnormality, marker chromosomes and triploidy, with particular emphasis on those such as microdeletions, unbalanced structural rearrangements and mosaicism which pose difficulties in detection by either array-CGH or cytogenetic examination. DNA was isolated from prenatal cultured amniocytes, cultured chorionic villi or postnatal bloods for samples previously confirmed as carrying a chromosomal rearrangement by either microscopic karyotype analysis or FISH. The results of these investigations were blinded prior to further analysis by array-CGH. Clones exceeding experimental thresholds were identified using methods as described earlier. http://jmg.bmj.com/

Uncultured samples Prior to array-CGH all samples were analysed by QF-PCR and conventional cytogenetic analysis, the results of which were concealed from the person performing array-CGH. For array-CGH, DNA was extracted from chorionic villi (a single small fragment) or 1-

2ml uncultured amniotic fluid using a Qiagen QIAamp blood DNA extraction mini kit on September 25, 2021 by guest. Protected copyright. with slight modification to the manufacturer’s protocol. Briefly, cell pellets were resuspended in 200µl PBS and the final elution volume was 100µl. As previously described, clones exceeding experimental thresholds were identified. All samples retrieved from patients undergoing diagnostic procedures, were surplus to diagnostic requirements, and were consented for use in research studies.

RESULTS Array validation Chromosome ‘add-in’ experiments To comprehensively test hybridisation characteristics of the entire clone set represented on the prenatal array, chromosome ‘add-in’ experiments were performed by the addition of DNA isolated from flow-sorted chromosomes to a series of self-versus-self hybridisations. Individual chromosomes were flow-sorted, with the exception of

6 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from chromosomes 9-12. Due to their similar size and base pair composition, these chromosomes are difficult to resolve into distinct populations and were therefore co- sorted (reviewed in28). A representative genomic profile for a chromosome 13 ‘add-in’ experiment is shown in fig 2. All clones mapping to chromosome 13 showed the expected increased ratio in response to the additional copies of the chromosome. However, 3 clones mapping to chromosomes 18 (RP11-16L7), 20 (RP13-329D4) and 22 (AC008103) also showed a significant ratio change identifying cross homology with chromosome 13 or incorrect mapping. Using this method for all chromosomes, we identified 69 incorrectly mapped or cross-hybridising clones (~10% of total clones), which either did not respond to the expected chromosome ‘add-in’, responded to DNA from a different chromosome or hybridised to multiple chromosomes. Clones identified using this method were excluded from all subsequent analyses and would not be included in future arrays.

Normal controls Validation experiments were performed in which DNA isolated from normal, healthy blood donors was hybridised against reference DNA and clones whose log2 ratios (test/reference) exceeded the thresholds set for each particular experiment were identified. In 30 normal-versus-normal reference hybridisations, 2 clones exceeded set thresholds. The log2 values for these clones were +0.37 and +0.38, which although significant, were substantially less than the theoretical ratios of +0.58 expected to indicate a single copy gain. Using the thresholds discussed earlier we would expect around 1 false positive result in every 24 hybridisations due to statistical variation. It would however be necessary to perform replicate or dye-swap hybridisations and to carry out alternative investigations such as FISH to distinguish a false-positive result from the possibility that these clones are reporting variation between normal individuals.

Patient samples http://jmg.bmj.com/ Array-CGH was performed using both a 1Mb resolution array and a custom-designed prenatal array in a blind study of 30 cultured prenatal or postnatal samples with known unbalanced karyotypes. Of the 30 aberrations, 22 were detected using the 1Mb resolution array, whereas 29 aberrations were identified using the prenatal array (table 1).

Table 1 on September 25, 2021 by guest. Protected copyright. Summary of cytogenetic analyses and array-CGH data using the prenatal and 1Mb microarrays on cultured prenatal and postnatal blood samples Array-CGH Sample Prenatal array 1Mb array Cytogenetic analysis 1 Deletion of 3 clones at 7q11.23 Normal 46,XY.ish del(7)(q11.23q11.23)(ELN-)*# 2 Deletion of 10 clones at Deletion of 7 clones at 46,XYdel(15)(q11.2q13).ish 15q11.2-15q13.1 15q11.2-15q13.1 del(15)(q11.2q11.2)(E24-)*# 3 Deletion of 10 clones at Deletion of 7 clones at 46,XX,del(15)(q11q13).ish 15q11.2-15q13.1 15q11.2-15q13.1 del(15)(q11q13)(D15S10-)*# 4 Deletion of 10 clones at Deletion of 7 clones at 46,XY,del(15)(q11.2q13).ish 15q11.2-15q13.1 15q11.2-15q13.1 del(15)(D15S10-)*# 5 Trisomy 21 Trisomy 21 47, XY, +21§± 6 Trisomy 21 Trisomy 21 47, XX, +21§± 7 Normal Normal 69, XXX§± 8 Trisomy 18 Trisomy 18 47, XX, +18§± 9 Dup(13ptel- Dup(13ptel- 46, XX, psu idic(13)(q32) §± 13q31.3),del(13q32.3-13qtel) 13q32.1),del(13q32.1-

7 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

13qtel) 10 Monosomy X Monosomy X 45, X§± 11 Trisomy 21 Trisomy 21 47, XX, +21§± 12 Dup(11q24.1- Dup(11q23.3- 46, XY, add(22)(q13.3).ish 11qtel),del(22q13.2-22qtel) 11qtel),del(22q13.31-22qtel) der(22)t(11;22)(q23;q13.3)(pVYS207M-, wcp11+,pVYS239A+)de novo§± 13 Deletion at 22q11.21 Normal 46, XY.ish del(22)(q11.2q11.2)(D22S931-, D22S941-)§# 14 Trisomy 13 Trisomy 13 47, XY, +13§± 15 Deletion of 3 clones at 7q11.23 Normal 46, XX.ish del (7)(q11.23q11.23)(ELN- ,LIMK1-,D7S613-) *# 16 Deletion of 7 clones at 17p11.2 Normal 46,XX,del(17)(p11.2p11.2).ish del(17)(p11.2p11.2)(D17S29-)*# 17 Deletion of 10 clones at 4ptel- Normal 46,XX.ish del(4)(p16.3)(WHSCR-)*# 4p16.3 18 Deletion at 22q11.21 Normal 46,XX,del(22)(q11.2q11.2).ish del(22)(q11.2q11.2)(TUPLE 1-)*# 19 Deletion at 22q11.21 Normal 46,XX.ish del(22)(q11.2q11.2)(TUPLE1-)*# 20 XXY XXY 47, XXY§± 21 Trisomy 13 Trisomy 13 46, XY, der(13;13)(q10;q10),+13§± 22 Duplication of 1 clone at Duplication of 8 clones at 47,XY,+mar.ish 12p11.21, ratios 12p11.22-12p11.1, ratios der(12)(wcp12+,D12Z3+)/46, XY§± indicative of less than a single indicative of less than a copy gain single copy gain 23 Monosomy X Monosomy X 45, X§± 24 Trisomy 18 Trisomy 18 47, XX, +18§± 25 XYY XYY 47, XYY§± 26 XXY XXY 47, XXY§± 27 Deletion of 3 clones at Deletion of 14 clones at 46, X, +mar.ish der (Y)(wcpY+)[25]/45, Yq11.221-Yq12 Yq11.221-Yq12 X[19] §± 28 Ratios for chromosome 9 clones Ratios for chromosome 9 47, XX, +9[68]/46, XX[5] §± show values indicative of less clones show values than a single copy gain indicative of less than a single copy gain 29 Trisomy 13 Trisomy 13 47, XX, +13§± 30 Deletion of 13 clones at Deletion of 17 clones at 46, XY, del(5)(p15.1) §± http://jmg.bmj.com/ 5p15.33-5p15.2 5p15.33-5p15.1 * Postnatal sample § Prenatal sample # Performed by FISH ± Performed by microscopic karyotype analysis

on September 25, 2021 by guest. Protected copyright. Eight cases of autosomal trisomy (chromosomes 13, 18 and 21(fig 3, A)), 5 examples of sex chromosome copy number abnormality (monosomy X, XXY and XYY) and microdeletions previously identified by FISH at 15q11 (Angelman and Prader-Willi syndromes, fig 3, B) and 5p (Cri-du-Chat syndrome) were observed by array-CGH using both arrays. However, microdeletions detected using the prenatal array at 4p (Wolf- Hirschhorn syndrome), 7q11 (Williams syndrome, fig 3, C), 22q11 (DiGeorge syndrome) and 17p11 (Smith-Magenis syndrome) were not observed using the 1Mb array as these regions are not covered by clones representing the 1Mb array clone-set.

Details of specific cases are given below Case 9 Microscopic karyotype analysis revealed an isodicentric chromosome 13, leading to trisomy for 13ptel-13q32 and monosomy for 13q32-13qtel. Array-CGH identified clones with ratios indicative of a single copy duplication or deletion.

8 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Case 12 Case 12 had an unbalanced translocation between chromosomes 11 and 22, leading to trisomy for 11q23-11qtel and monosomy for the terminal region of the long arm of chromosome 22. Case 22 Karyotype analysis showed 34/40 cells having an additional, small marker chromosome, positive for a chromosome 12 centromere repeat probe and a whole chromosome 12 FISH paint. Array-CGH using the prenatal array detected duplicated ratios for a single chromosome 12 clone, although the hybridisation ratios did not reach the values expected for a single copy gain, indicating either partial clone duplication or a mosaic karyotype. A similar result was observed with 12 chromosome 12 clones using the 1Mb resolution array. Case 27 Case 27 showed 25/44 cells having a marker chromosome, which was positive for a whole Y chromosome FISH paint. This was identified using array analysis by the deletion of 3 or 14 clones (prenatal and 1Mb array respectively). Case 28 Case 28 also exhibited a mosaic karyotype with 68/73 cells having an extra chromosome 9. Clones on chromosome 9 showed increased ratios by array-CGH, but less than those expected for a single copy gain. Case7 Neither array was able to identify an abnormality in case 7, which had a triploid karyotype.

Uncultured prenatal samples Array-CGH was performed using the prenatal array in a blind study of 30 uncultured prenatal samples. Samples had been previously analysed by QF-PCR and by microscopic http://jmg.bmj.com/ karyotype analysis (publication in progress). They included 20 normal controls, 8 cases of trisomy 21, one case of triploidy with 4 copies of chromosomes 5 and 21, and a 45, X/ 46, X dic Yp mosaic. Representative hybridisations are shown in figure 4. In 29/30 cases, there was concordance between array-CGH and previous analyses (table 2).

Table 2 on September 25, 2021 by guest. Protected copyright. Summary of cytogenetic analyses, QF-PCR and array-CGH data using the prenatal microarray on uncultured prenatal samples Sample Prenatal array Karyotype as ascertained by Concordance between conventional karyotyping array-CGH, QF-PCR and karyotype 31# Normal female 46, XX Yes 32# Trisomy 21 47, XY +21 Yes 33# Normal female 46, XX Yes 34# Normal female 46, XX Yes 35# Normal male 46, XY Yes 36# Normal male 46, XY Yes 37# Normal male 46, XY Yes 38$ Normal male 46, XY Yes 39# Duplication of 5 clones at Yptel- 45, X/ 46, X dic Yp mosaic Yes Yq11.223 40# Normal female 46, XX Yes 41# Normal female 46, XX Yes

9 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

42# Normal male 46, XY Yes 43# Normal female 46, XX Yes 44# Trisomy 21 47, XY +21 Yes 45# Trisomy 21 47, XY +21 Yes 46# Normal female 46, XX Yes 47# Normal male 46, XY Yes 48# Trisomy 21 47, XY +21 Yes 49# Trisomy 21 47, XY +21 Yes 50# Ratios indicative of single copy gain 71, XXY +5 +21 No for chromosomes 5 and 21 51# Trisomy 21 47, XY +21 Yes 52# Trisomy 21 47, XY +21 Yes 53# Normal female 46, XX Yes 54# Normal female 46, XX Yes 55# Normal male 46, XY Yes 56# Normal male 46, XY Yes 57# Trisomy 21 47, XY +21 Yes 58# Normal male 46, XY Yes 59# Normal male 46, XY Yes 60# Normal male 46, XY Yes #Amniotic fluid sample $CVS sample

Array-CGH failed to detect triploidy for case 50, but this sample was identified as being abnormal as ratios for clones representing chromosomes 5 and 21 were indicative of a single copy gain.

DISCUSSION We have demonstrated the potential utility of array-CGH for the detection of chromosomal abnormalities in prenatal diagnosis. By comparing established methods of prenatal testing (microscopic karyotype analysis, FISH and QF-PCR) with array-CGH, using both a small, custom-designed array and a larger array containing clones covering http://jmg.bmj.com/ the entire genome at an approximate resolution of 1Mb, we were able to show that the great majority of abnormalities can be detected using array-based methods.

The 1Mb array used in this study was previously shown to be successful for the detection 16 of submicroscopic deletions in patients with mental retardation/learning difficulties . In on September 25, 2021 by guest. Protected copyright. the current study, this array detected only 22/30 known aberrations, missing small microdeletions at 4p (Wolf-Hirshhorn syndrome), 7q (Williams syndrome), 17p (Smith- Magenis syndrome) and 22q (DiGeorge syndrome) because of low clone density in these regions.

In contrast, the smaller, purpose designed prenatal array identified 29 of the 30 aberrations. This lower-density array is comprised of clones for genome-wide analysis at an approximate resolution of only around 10Mb (~450 clones) but with an additional set of around 200 clones, chosen to cover major trisomy and microdeletion regions at increased density. This array, designed with a lower overall complexity but containing clinically important clones, identified all but one of the known chromosomal abnormalities.

10 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Particularly in light of emerging information about large copy-number variation in normal human populations21, 22, large arrays are likely to generate some difficulties in interpretation, at least until much more is learned about genomic copy number variation. Small, low resolution arrays are likely to be less prone to technical error, produce fewer false positive results, will be cheaper to standardise and fabricate and can be designed to avoid genomic regions with known polymorphic copy number variation. This and earlier studies18 demonstrate that a targeted array can combine these advantages with a high detection rate for the chosen diagnostic categories.

It is important to address some of the limitations of array-CGH before this technique is considered for clinical diagnosis (reviewed in29, 30). Array-CGH is unable to detect polyploidy and balanced chromosomal rearrangements as the process of data normalisation produces results only sensitive to relative dosage imbalance between different regions of the sample under test. Polyploidy (almost always triploidy) is virtually always lethal during foetal life and is generally suspected on ultrasound investigation. Where such suspicion arises, other diagnostic procedures such as QF-PCR must be used. While conventional karyotyping is able to detect balanced chromosomal rearrangements, the majority of truly balanced translocations produce no phenotypic abnormality31 and their discovery can lead to difficult clinical decisions during pregnancy. Detection of copy number variation by array-CGH does not itself allow non- disjunction to be distinguished from unbalanced structural abnormality which is important for genetic counselling.

On the other hand, array-CGH using a carefully selected clone-set is likely to be more sensitive than microscopic karyotyping for detecting microdeletions in the tested regions. Microdeletion syndromes are contiguous gene syndromes characterised by the loss of specific chromosomal segments, and result in clinically distinct phenotypes, some of http://jmg.bmj.com/ which are at least as severe as those observed with Down syndrome (reviewed in32). Many of the chromosomal rearrangements associated with microdeletion syndromes are too small to be reliably identified by routine microscopic karyotype analysis and are therefore often only detected after birth by FISH analysis using locus specific probes33-35.

Precisely which chromosomal regions to include in a designed prenatal array will require on September 25, 2021 by guest. Protected copyright. continued discussion including laboratory, clinical and patient viewpoints. In the array described in this study, we included major eponymous microdeletion regions. However, we deliberately excluded regions where copy number change is associated with mild or no phenotypic abnormality. Inclusion of more pericentromeric clones would improve the likelihood of detecting supernumerary marker chromosomes.

Array-CGH using a small array is therefore likely to detect all major aneuploidies, and to be more sensitive than microscopy at detecting microdeletions within areas of suspicion. It will not detect polyploidy, balanced structural rearrangements, and some unexpected relatively small unbalanced rearrangements which happen not to fall over areas of clone coverage. Mosaicism, by its nature, is likely to present problems for either technique. In this series we noted clone ratios suggestive of less than a single copy gain in 2 of the 3 mosaics tested, with the third case of mosaicism being identified as a deletion, but further

11 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from data is needed to define the real sensitivity of array-CGH for mosaicism. In practise, array-CGH (in common with QF-PCR and FISH) will be used on uncultured material, and is therefore likely to produce some differences from cytogenetics performed on cultured cells with the opportunity for differential cell lineage proliferation. We would estimate that, over all categories of abnormality, the two techniques will have rather similar sensitivity for detection of clinically important changes.

Array-CGH analysis of cell-free foetal DNA isolated from amniotic fluid samples has shown promising results, with foetal sex and whole-chromosome changes being determined36. However, higher levels of clone to clone variability were noted when compared to cultured amniocytes, so it is possible that smaller changes involving only a few clones might not be reliably identified using this method. In this study, DNA isolated from both cultured and uncultured amniocytes was used for hybridisation. We found that array-CGH can work well on DNA from uncultured amniocytes. Since we have had to work with material surplus to diagnostic requirement, we performed array-CGH on DNA derived from as little as 1ml of uncultured amniotic fluid; in practice, performance would doubtless be better with more material. The ability to perform array-CGH on uncultured prenatal cells has not previously been established.

Array-CGH has the potential to combine the speed and capacity for automation of a DNA based technique, with the genome-wide scanning capability of conventional cytogenetics. We believe these results provide proof of principle that array-CGH utilising a purpose designed array is rapid, and may have about the same overall capacity to detect clinically relevant chromosome abnormalities as conventional cytogenetics. It has the potential to replace karyotyping for prenatal cytogenetics, but considerable further work is required to reach a consensus on the optimum configuration of an array for clinical use; and large trials are needed to demonstrate sensitivity and specificity in clinical operating conditions http://jmg.bmj.com/ before the clinical implementation of array-CGH can be considered.

ACKNOWLEDGEMENTS We would like to thank Lucy Raymond and Josep Parnau for the normal control samples used in this study. We are grateful to Christine Hall for providing DNA isolated from prenatal amniocytes and to Ingrid Simonic for many useful clinical discussions. We on September 25, 2021 by guest. Protected copyright. would like to acknowledge Cordelia Langford and the Wellcome Trust Sanger Institute Microarray Facility for printing the arrays. This work was supported in part by the Wellcome Trust.

Figure legends

Figure 1 Ideogram showing the approximate locations for clones represented on the prenatal microarray

Figure 2 Chromosome 13 ‘add-in’ experiment, demonstrating the response of clones represented on the prenatal array to additional copies of chromosome 13. (A) Shows the response of

12 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from all clones to approximately 2 additional copies of chromosome 13, whereas (B) shows the response of chromosome 13 clones only to approximately 2 or 4 additional copies of chromosome 13

Figure 3 Array-CGH data for samples 5(A), 1(B) and 2(C) (table 1), on both the prenatal (a) and 1Mb (b) arrays. The X axis represents the chromosome (A) or the distance in Mb from the p telomere (B and C). The Y axis represents the hybridisation ratio given as a log2 scale. Significant gains are shown in green whereas significant losses are red (A) trisomy 21, (B) deletion at 7q11.23 observed on the prenatal array but absent at 1Mb resolution and (C) deletion at 15q11.2-15q13.1 visible on both arrays

Figure 4 Array-CGH data for samples 35(A), 46(B), 52(C) and 39(D) (table 2), on the prenatal array. The X axis represents the chromosome (A, B, and C) or the distance in Mb from the p telomere (D). The Y axis represents the hybridisation ratio given as a log2 scale. Significant gains are shown in green whereas significant losses are red. (A) 46, XY, (B) 46, XX (C) 47, XY +21 and (D) 46, XY +duplication of 5 clones at Yptel-Yq11.223

http://jmg.bmj.com/

REFERENCES 1 Waters JJ, Waters KS. Trends in cytogenetic prenatal diagnosis in the UK: results

from UKNEQAS external audit, 1987-1998. Prenat Diagn 1999;19(11):1023-6. on September 25, 2021 by guest. Protected copyright. 2 Caspersson T, Farber S, Foley GE, Kudynowski J, Modest EJ, Simonsson E, Wagh U, Zech L. Chemical differentiation along metaphase chromosomes. Exp Cell Res 1968;49(1):219-22. 3 Shaffer LG, Bejjani BA. A cytogeneticist's perspective on genomic microarrays. Hum Reprod Update 2004;10(3):221-6. 4 Shaffer LG, Lupski JR. Molecular mechanisms for constitutional chromosomal rearrangements in humans. Annu Rev Genet 2000;34:297-329. 5 Cirigliano V, Voglino G, Canadas MP, Marongiu A, Ejarque M, Ordonez E, Plaja A, Massobrio M, Todros T, Fuster C, Campogrande M, Egozcue J, Adinolfi M. Rapid prenatal diagnosis of common chromosome aneuploidies by QF-PCR. Assessment on 18,000 consecutive clinical samples. Mol Hum Reprod 2004;10(11):839-46.

13 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

6 Mann K, Fox SP, Abbs SJ, Yau SC, Scriven PN, Docherty Z, Ogilvie CM. Development and implementation of a new rapid aneuploidy diagnostic service within the UK National Health Service and implications for the future of prenatal diagnosis. Lancet 2001;358(9287):1057-61. 7 Mansfield ES. Diagnosis of Down syndrome and other aneuploidies using quantitative polymerase chain reaction and small tandem repeat polymorphisms. Hum Mol Genet 1993;2(1):43-50. 8 Klinger K, Landes G, Shook D, Harvey R, Lopez L, Locke P, Lerner T, Osathanondh R, Leverone B, Houseal T, et al. Rapid detection of chromosome aneuploidies in uncultured amniocytes by using fluorescence in situ hybridization (FISH). Am J Hum Genet 1992;51(1):55-65. 9 Ryall RG, Callen D, Cocciolone R, Duvnjak A, Esca R, Frantzis N, Gjerde EM, Haan EA, Hocking T, Sutherland G, Thomas DW, Webb F. Karyotypes found in the population declared at increased risk of Down syndrome following maternal serum screening. Prenat Diagn 2001;21(7):553-7. 10 Kallioniemi A, Kallioniemi OP, Sudar D, Rutovitz D, Gray JW, Waldman F, Pinkel D. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumors. Science 1992;258(5083):818-21. 11 Albertson DG, Pinkel D. Genomic microarrays in human genetic disease and cancer. Hum Mol Genet 2003;12 Spec No 2:R145-52. 12 Daniely M, Aviram-Goldring A, Barkai G, Goldman B. Detection of chromosomal aberration in fetuses arising from recurrent spontaneous abortion by comparative genomic hybridization. Hum Reprod 1998;13(4):805-9. 13 Kirchhoff M, Pedersen S, Kjeldsen E, Rose H, Duno M, Kolvraa S, Lundsteen C. Prospective study comparing HR-CGH and subtelomeric FISH for investigation of individuals with mental retardation and dysmorphic features and an update of a study using only HR-CGH. Am J Med Genet A 2004;127(2):111-7. http://jmg.bmj.com/ 14 Pinkel D, Segraves R, Sudar D, Clark S, Poole I, Kowbel D, Collins C, Kuo WL, Chen C, Zhai Y, Dairkee SH, Ljung BM, Gray JW, Albertson DG. High resolution analysis of DNA copy number variation using comparative genomic hybridization to microarrays. Nat Genet 1998;20(2):207-11. 15 Ishkanian AS, Malloff CA, Watson SK, DeLeeuw RJ, Chi B, Coe BP, Snijders A,

Albertson DG, Pinkel D, Marra MA, Ling V, MacAulay C, Lam WL. A tiling on September 25, 2021 by guest. Protected copyright. resolution DNA microarray with complete coverage of the human genome. Nat Genet 2004;36(3):299-303. 16 Shaw-Smith C, Redon R, Rickman L, Rio M, Willatt L, Fiegler H, Firth H, Sanlaville D, Winter R, Colleaux L, Bobrow M, Carter NP. Microarray based comparative genomic hybridisation (array-CGH) detects submicroscopic chromosomal deletions and duplications in patients with learning disability/mental retardation and dysmorphic features. J Med Genet 2004;41(4):241-8. 17 Vissers LE, de Vries BB, Osoegawa K, Janssen IM, Feuth T, Choy CO, Straatman H, van der Vliet W, Huys EH, van Rijk A, Smeets D, van Ravenswaaij-Arts CM, Knoers NV, van der Burgt I, de Jong PJ, Brunner HG, van Kessel AG, Schoenmakers EF, Veltman JA. Array-based comparative genomic hybridization

14 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

for the genomewide detection of submicroscopic chromosomal abnormalities. Am J Hum Genet 2003;73(6):1261-70. 18 Bejjani BA, Saleki R, Ballif BC, Rorem EA, Sundin K, Theisen A, Kashork CD, Shaffer LG. Use of targeted array-based CGH for the clinical diagnosis of chromosomal imbalance: is less more? Am J Med Genet A 2005;134(3):259-67. 19 Le Caignec C, Boceno M, Saugier-Veber P, Jacquemont S, Joubert M, David A, Frebourg T, Rival JM. Detection of genomic imbalances by array based comparative genomic hybridisation in fetuses with multiple malformations. J Med Genet 2005;42(2):121-8. 20 Schaeffer AJ, Chung J, Heretis K, Wong A, Ledbetter DH, Lese Martin C. Comparative genomic hybridization-array analysis enhances the detection of aneuploidies and submicroscopic imbalances in spontaneous miscarriages. Am J Hum Genet 2004;74(6):1168-74. 21 Iafrate AJ, Feuk L, Rivera MN, Listewnik ML, Donahoe PK, Qi Y, Scherer SW, Lee C. Detection of large-scale variation in the human genome. Nat Genet 2004;36(9):949-51. 22 Sebat J, Lakshmi B, Troge J, Alexander J, Young J, Lundin P, Maner S, Massa H, Walker M, Chi M, Navin N, Lucito R, Healy J, Hicks J, Ye K, Reiner A, Gilliam TC, Trask B, Patterson N, Zetterberg A, Wigler M. Large-scale copy number polymorphism in the human genome. Science 2004;305(5683):525-8. 23 Fiegler H, Carr P, Douglas EJ, Burford DC, Hunt S, Scott CE, Smith J, Vetrie D, Gorman P, Tomlinson IP, Carter NP. DNA microarrays for comparative genomic hybridization based on DOP-PCR amplification of BAC and PAC clones. Genes Chromosomes Cancer 2003;36(4):361-74. 24 Gilbert N, Boyle S, Fiegler H, Woodfine K, Carter NP, Bickmore WA. Chromatin architecture of the human genome: gene-rich domains are enriched in open chromatin fibers. Cell 2004;118(5):555-66. http://jmg.bmj.com/ 25 Jain AN, Tokuyasu TA, Snijders AM, Segraves R, Albertson DG, Pinkel D. Fully automatic quantification of microarray image data. Genome Res 2002;12(2):325- 32. 26 Carter NP. Bivariate chromosome analysis using a commercial flow cytometer. Methods Mol Biol 1994;29:187-204.

27 Fiegler H, Gribble SM, Burford DC, Carr P, Prigmore E, Porter KM, Clegg S, on September 25, 2021 by guest. Protected copyright. Crolla JA, Dennis NR, Jacobs P, Carter NP. Array painting: a method for the rapid analysis of aberrant chromosomes using DNA microarrays. J Med Genet 2003;40(9):664-70. 28 Ibrahim SF, van den Engh G. High-speed chromosome sorting. Chromosome Res 2004;12(1):5-14. 29 Vermeesch JR, Melotte C, Froyen G, Van Vooren S, Dutta B, Maas N, Vermeulen S, Menten B, Speleman F, De Moor B, Van Hummelen P, Marynen P, Fryns JP, Devriendt K. Molecular karyotyping: array CGH quality criteria for constitutional genetic diagnosis. J Histochem Cytochem 2005;53(3):413-22. 30 Bejjani BA, Theisen AP, Ballif BC, Shaffer LG. Array-based comparative genomic hybridization in clinical diagnosis. Expert Rev Mol Diagn 2005;5(3):421-9.

15 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

31 Warburton D. De novo balanced chromosome rearrangements and extra marker chromosomes identified at prenatal diagnosis: clinical significance and distribution of breakpoints. Am J Hum Genet 1991;49(5):995-1013. 32 Malcolm S. Microdeletion and microduplication syndromes. Prenat Diagn 1996;16(13):1213-9. 33 Juyal RC, Figuera LE, Hauge X, Elsea SH, Lupski JR, Greenberg F, Baldini A, Patel PI. Molecular analyses of 17p11.2 deletions in 62 Smith-Magenis syndrome patients. Am J Hum Genet 1996;58(5):998-1007. 34 Wu YQ, Sutton VR, Nickerson E, Lupski JR, Potocki L, Korenberg JR, Greenberg F, Tassabehji M, Shaffer LG. Delineation of the common critical region in Williams syndrome and clinical correlation of growth, heart defects, ethnicity, and parental origin. Am J Med Genet 1998;78(1):82-9. 35 Zollino M, Di Stefano C, Zampino G, Mastroiacovo P, Wright TJ, Sorge G, Selicorni A, Tenconi R, Zappala A, Battaglia A, Di Rocco M, Palka G, Pallotta R, Altherr MR, Neri G. Genotype-phenotype correlations and clinical diagnostic criteria in Wolf-Hirschhorn syndrome. Am J Med Genet 2000;94(3):254-61. 36 Larrabee PB, Johnson KL, Pestova E, Lucas M, Wilber K, LeShane ES, Tantravahi U, Cowan JM, Bianchi DW. Microarray analysis of cell-free fetal DNA in amniotic fluid: a prenatal molecular karyotype. Am J Hum Genet 2004;75(3):485-91. 37 Cheung SW, Shaw CA, Yu W, Li J, Ou Z, Patel A, Yatsenko SA, Cooper ML, Furman P, Stankiewicz P, Lupski JR, Chinault AC, Beaudet AL. Development and validation of a CGH microarray for clinical cytogenetic diagnosis. Genet Med 2005;7(6):422-32.

http://jmg.bmj.com/ on September 25, 2021 by guest. Protected copyright.

16 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of all authors, an exclusive licence (or non exclusive for government employees) on a worldwide basis to the BMJ Publishing Group Ltd and its Licensees to permit this article (if accepted) to be published in JMG editions and any other BMJPGL products to exploit all subsidiary rights, as set out in our licence (http://jmg.bmjjournals.com/misc/ifora/licenceform.shtml) http://jmg.bmj.com/ on September 25, 2021 by guest. Protected copyright.

17 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from http://jmg.bmj.com/ on September 25, 2021 by guest. Protected by copyright. J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from http://jmg.bmj.com/ on September 25, 2021 by guest. Protected by copyright. J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from http://jmg.bmj.com/ on September 25, 2021 by guest. Protected by copyright. J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from http://jmg.bmj.com/ on September 25, 2021 by guest. Protected by copyright. J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Position Position Position Clone ID Chr (Mb) Clone ID Chr (Mb) Clone ID Chr (Mb) RP4-575L21 1 10.13 RP11-397D2 2 223.05 RP11-501E14 4 121.95 RP4-745E8 1 20.37 RP11-149O3 2 228.57 RP11-69O1 4 141.83 RP5-893G23 1 29.56 RP11-497D24 2 238.37 RP11-384D10 4 172.43 RP11-204L3 1 40.04 RP11-463B12 2 240.72 RP11-226A18 4 182.24 RP1-86A18 1 50.64 PAC1186B18 3 0.24 RP11-354H17 4 190.19 RP11-37M11 1 59.95 RP11-86C13 3 0.86 RP11-20K9 5 3.26 RP5-944F13 1 69.82 RP11-94A14 3 10.70 RP11-103L11 5 3.91 RP11-413E1 1 89.87 RP11-180N14 3 21.40 RP11-334G8 5 8.16 RP11-143H12 1 98.70 RP11-7I16 3 30.70 RP11-215I16 5 8.27 RP11-413P11 1 99.69 RP11-527M19 3 41.26 RP11-480D4 5 8.47 RP11-260K3 1 99.69 RP11-89F17 3 51.56 RP11-143A12 5 9.01 RP4-667F15 1 109.45 RP11-154D3 3 62.10 RP11-5H13 5 9.23 RP11-418J17 1 119.52 RP11-522N9 3 72.54 RP11-193P20 5 9.37 RP1-13P20 1 149.56 RP11-206J21 3 82.73 RP11-32D12 5 9.52 RP11-430G6 1 159.77 RP11-88I7 3 95.61 RP11-447B18 5 9.92 RP11-277C14 1 168.66 RP11-40M23 3 105.26 RP11-54F2 5 10.59 RP11-71D4 1 179.44 RP11-107J18 3 115.54 RP11-360C3 5 11.85 RP11-239J11 1 189.94 RP11-9N20 3 124.94 RP11-269G2 5 13.30 RP11-480I12 1 199.54 RP11-165M11 3 144.94 RP11-19O2 5 16.28 RP11-354K1 1 208.20 RP1-38G8 3 145.31 RP11-61K20 5 20.12 RP11-308L13 1 218.20 RP11-223L18 3 155.03 RP11-259G23 5 22.39 RP11-99J16 1 227.30 RP11-90M7 3 164.44 RP11-116O11 5 24.61 RP11-553N16 1 238.36 RP11-569P10 3 174.46 RP11-192H6 5 25.26 RP11-438H8 1 244.36 RP11-14I2 3 184.90 RP11-351N6 5 26.45 CTB-160H23 1 245.34 RP11-47G2 3 193.83 RP11-461E7 5 32.32 GS1-68F18 2 0.31 RP11-252K11 3 197.40 RP11-110H4 5 34.22 GS1-8L3 2 0.38 CTC-196F4 3 198.47 RP11-79C6 5 35.91

RP11-352J11 2 2.29 CTC-36P21 4 0.13 RP11-117M17 5 39.64 http://jmg.bmj.com/ RP11-333O1 2 12.84 RP11-386I15 4 1.27 RP11-112L7 5 42.17 RP11-368O18 2 22.58 RP11-572O17 4 1.69 RP11-575G10 5 42.90 RP11-559D11 2 32.62 RP11-317B7 4 2.09 RP11-92M7 5 52.99 RP11-299C5 2 42.48 RP11-478C1 4 2.22 RP11-18K15 5 63.41 RP11-389K20 2 52.20 RP11-478A6 4 2.49 RP11-97L2 5 73.82

RP11-422B1 2 58.21 RP11-520M5 4 3.03 RP11-72L22 5 86.42 on September 25, 2021 by guest. Protected copyright. RP11-52F10 2 62.73 RP11-357G3 4 3.38 RP11-526D16 5 95.92 RP11-343N14 2 72.40 RP11-323F5 4 4.69 RP11-319P13 5 106.83 RP11-495B16 2 82.69 RP11-386I19 4 5.13 RP11-434D11 5 126.14 RP11-451G1 2 103.99 RP11-183A12 4 14.76 RP11-114H21 5 135.92 RP11-528G9 2 110.26 RP11-339D20 4 19.75 RP11-481E16 5 146.40 RP11-412A2 2 115.86 RP11-10G12 4 22.30 RP11-102A1 5 157.26 RP11-414I8 2 137.57 RP11-559P10 4 24.66 RP11-281O15 5 178.35 RP11-3P4 2 148.13 RP11-390C19 4 29.54 CTC-240G13 5 180.57 RP11-125M15 2 167.66 RP11-143G24 4 34.80 CTB-62I11 6 0.15 RP11-157E8 2 177.25 RP11-24I7 4 62.57 RP11-15N12 6 3.38 RP11-123G24 2 188.40 RP11-1J11 4 72.47 RP11-97A19 6 12.49 RP11-172B9 2 197.88 RP11-263F19 4 83.40 RP11-33I5 6 22.66 RP11-23F11 2 200.23 RP11-11N6 4 91.98 RP11-175A4 6 33.54 RP11-309L6 2 207.89 RP11-167N19 4 102.03 RP11-227E22 6 44.06 RP11-423F9 2 218.02 RP11-326N15 4 112.55 RP11-362K18 6 54.17 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Position Position Position Clone ID Chr (Mb) Clone ID Chr (Mb) Clone ID Chr (Mb) RP11-349P19 6 65.19 RP11-373L22 8 102.63 RP11-125G9 12 20.89 RP11-256L9 6 73.19 RP11-544H16 8 112.60 RP11-77I22 12 30.84 RP11-25O6 6 83.50 RP11-16G11 8 122.80 RP11-549E9 12 39.75 RP11-21G12 6 93.15 RP11-71N3 8 133.24 RP11-112N23 12 49.11 RP11-368P1 6 142.29 RP11-370K2 8 142.70 RP11-410B16 12 60.29 RP11-450E24 6 152.28 CTC-489D14 8 146.08 RP11-101K2 12 69.82 RP11-288H12 6 160.55 RP11-48M17 9 2.22 RP11-268A19 12 80.32 RP1-137D17 6 169.61 RP11-446F13 9 12.31 RP11-12K13 12 82.75 CTB-57H24 6 170.75 RP11-495L19 9 23.47 RP11-239F20 12 90.44 CTB-164D18 7 0.22 RP11-274B18 9 68.44 RP11-285E23 12 100.65 RP11-106E3 7 2.58 RP11-174K23 9 78.63 RP11-438N16 12 112.72 RP11-195L14 7 12.81 RP11-176L21 9 88.72 RP11-338K17 12 122.73 RP11-451F11 7 23.55 RP11-92C4 9 98.72 RP11-46H11 12 132.05 RP11-36H20 7 43.32 RP11-400A24 9 108.37 CTC-221K18 12 132.29 RP11-449G3 7 54.50 RP11-451E16 9 118.21 RP11-76K19 13 19.14 RP11-458F8 7 65.82 RP11-545E17 9 128.62 RP11-218E6 13 28.92 RP11-32N3 7 71.45 RP11-399H11 9 135.31 RP11-407E23 13 39.50 RP11-535E8 7 71.58 GS1-135I17 9 138.22 RP11-185C18 13 48.99 RP11-483G21 7 72.16 CTC-306F7 10 0.31 RP11-359P14 13 59.78 RP4-665P5 7 73.40 RP11-29A19 10 1.29 RP11-436I5 13 69.23 RP5-1186P10 7 73.49 RP11-401F24 10 11.91 RP11-421K11 13 80.81 RP11-451K15 7 74.50 RP11-165O3 10 21.32 RP11-388D4 13 89.39 RP4-754G14 7 74.55 RP11-472N13 10 31.95 RP11-118F16 13 100.40 RP11-107L23 7 74.87 RP11-38B21 10 43.94 RP11-17E4 13 110.60 RP5-1188N21 7 75.03 RP11-47O13 10 52.98 CTB-163C9 13 113.83 RP5-1129E22 7 75.70 RP11-809M12 10 62.95 RP11-98N22 14 19.65 RP11-467H10 7 76.40 RP11-95M17 10 84.50 RP11-159L20 14 30.24

RP11-275G11 7 76.56 RP11-348J12 10 94.82 RP11-332O9 14 40.46 http://jmg.bmj.com/ RP11-144P23 7 77.72 RP11-724N1 10 104.73 RP11-255G12 14 50.94 RP11-343J14 7 78.20 RP11-357H24 10 114.97 RP11-307P22 14 60.30 RP5-1057M1 7 79.40 RP11-436O19 10 124.18 RP11-486O13 14 69.70 RP11-212B1 7 86.79 RP11-45A17 10 133.70 RP11-526N18 14 79.43 RP11-443I10 7 107.28 RP11-496I9 11 0.58 RP11-79J20 14 88.90

RP11-374M7 7 116.21 RP11-327O2 11 10.37 RP11-123M6 14 100.38 on September 25, 2021 by guest. Protected copyright. RP11-224A1 7 126.99 RP11-6K5 11 20.29 RP11-13O24 15 20.26 RP11-102G17 7 136.50 RP11-302O8 11 30.65 RP11-494F2 15 21.53 RP11-302C22 7 146.98 RP11-227P3 11 50.01 RP11-350A1 15 21.90 RP11-452C13 7 157.22 RP11-286N22 11 60.96 RP11-385H1 15 22.56 CTB-3K23 7 158.37 RP11-598K3 11 70.31 RP11-131I21 15 22.82 BAC114J18 8 1.05 RP11-187P2 11 80.49 RP11-32L10 15 23.41 RP11-104F14 8 2.36 RP11-268B20 11 90.81 RP11-446P9 15 23.74 RP11-589N15 8 11.80 RP11-21G19 11 101.33 RP11-150C6 15 24.82 RP11-177H13 8 23.20 RP11-108O10 11 111.19 RP11-570N16 15 25.05 RP11-301H15 8 32.62 RP11-93E4 11 122.21 RP11-30G8 15 25.52 RP11-503E24 8 42.58 PAC1064E20 11 131.01 RP11-263I19 15 41.77 RP11-197I11 8 53.17 RP11-545G16 11 132.54 RP11-232J12 15 51.74 RP11-351E7 8 73.59 CTC-496A11 12 0.71 RP11-24N10 15 61.03 RP11-34M16 8 82.70 RP11-359B12 12 0.97 RP11-272D12 15 71.36 RP11-3J21 8 93.23 RP11-434C1 12 11.63 RP11-127F21 15 81.27 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Position Position Position Clone ID Chr (Mb) Clone ID Chr (Mb) Clone ID Chr (Mb) RP11-266O8 15 91.63 RP5-836L9 17 20.07 RP11-351D2 21 43.64 CTB-154P1 15 99.97 RP11-121A13 17 20.17 RP11-178H12 21 46.52 RP11-473M20 16 3.07 RP11-434D2 17 20.33 AC016026 22 16.71 RP11-433P17 16 3.47 RP11-344E13 17 20.70 AC016027 22 16.88 RP11-461A8 16 3.63 RP11-474K4 17 27.51 XX-bac32 22 16.96 RP11-95J11 16 3.88 RP11-156E6 17 37.28 AC008079 22 17.01 AC009171 16 4.20 RP11-248L3 17 48.51 AC007326 22 17.32 RP11-295D4 16 4.38 RP11-156L14 17 58.15 AC000095 22 17.39 RP11-35P16 16 4.52 RP11-171G2 17 68.28 AC004461 22 17.41 RP11-61L4 16 4.88 GS1-50C4 17 77.81 AC004462 22 17.45 RP11-490O6 16 11.80 RP11-567O16 17 78.37 AC000081 22 17.59 RP11-101E7 16 21.94 GS1-74G18 18 0.21 AC000094 22 17.63 RP11-147B17 16 49.17 RP11-324G2 18 0.25 AC000085 22 17.67 RP11-457D20 16 58.89 RP11-419J16 18 10.10 AC000092 22 17.74 RP11-311C24 16 68.17 RP11-17J14 18 19.10 AC000092 22 17.74 RP11-556H2 16 77.69 RP11-413I9 18 28.34 AC000079 22 17.78 RP11-21B21 16 87.17 RP11-215A20 18 58.66 AC000087 22 17.84 CTC-240G10 16 88.49 RP11-169F17 18 68.91 AC000088 22 17.88 CTB-121I4 16 88.55 RP11-154H12 18 75.62 AC000082 22 17.90 RP11-216P6 17 0.91 CTC-964M9 18 75.94 AC000086 22 17.97 RP11-4F24 17 1.57 RP11-500M22 19 4.91 AC000077 22 18.00 RP11-380H7 17 2.14 RP5-859H16 19 5.64 XX-91c 22 18.15 RP1-59D14 17 2.25 RP11-197O4 19 10.33 AC000089 22 18.17 RP11-135N5 17 2.40 CTD-2332E11 19 20.55 AC000089 22 18.17 RP11-74E22 17 2.56 RP11-359H18 19 23.64 AC000076 22 18.20 RP11-64J4 17 3.19 CTD-2527I21 19 40.37 AC000078 22 18.24 RP11-48B14 17 3.48 CTB-14D10 19 51.09 AC005663 22 18.36

RP11-167N20 17 3.75 RP11-423F16 19 56.41 AC006547 22 18.49 http://jmg.bmj.com/ RP11-104O19 17 4.09 RP11-394L10 19 61.00 AC007731 22 19.11 RP11-460M1 17 4.27 GS1-1129C9 19 63.69 AC005500 22 19.12 RP11-198F11 17 4.59 CTB-106I1 20 0.23 AC004033 22 19.21 RP11-401O9 17 10.15 RP5-852M4 20 0.35 AC002470 22 19.56 RP11-385D13 17 15.42 RP4-599I11 20 4.70 AC002472 22 19.71

RP11-138I1 17 16.21 RP11-204H22 20 10.49 D86995 22 20.60 on September 25, 2021 by guest. Protected copyright. RP1-77H15 17 16.36 RP5-822J19 20 15.56 D87019 22 20.62 RP11-92B11 17 16.56 RP1-167O22 20 21.54 D87012 22 20.65 RP11-416I2 17 16.88 RP5-1085F17 20 30.85 RP11-80O7 22 22.64 RP11-45M22 17 17.04 RP4-633O20 20 35.88 RP1-76B20 22 28.46 RP11-524F11 17 17.43 RP1-138B7 20 41.67 XXbac-677f7 22 31.56 RP1-253P7 17 17.67 RP11-347D21 20 46.06 RP1-215F16 22 33.81 RP11-258F1 17 17.95 RP13-379L11 20 56.02 RP1-172B20 22 38.45 RP11-34O10 17 18.23 CTB-81F12 20 62.39 RP3-355C18 22 39.54 RP11-158M20 17 18.30 RP1-126N20 21 14.68 RP3-388M5 22 42.56 RP11-28B23 17 18.76 RP11-304D2 21 18.30 RP11-406A18 X 21.37 RP11-135L13 17 19.10 RP11-25F24 21 23.57 RP5-1147O16 X 32.26 CTB-187M2 17 19.29 RP1-64G16 21 28.69 RP1-308O1 X 42.06 RP11-311F12 17 19.58 RP11-102E10 21 36.85 RP11-56H2 X 51.12 RP11-78O7 17 19.68 RP11-98O13 21 38.08 RP11-284B18 X 63.30 RP11-209D14 17 19.82 RP11-164E1 21 38.67 RP11-236O12 X 74.65 J Med Genet: first published as 10.1136/jmg.2005.037648 on 30 September 2005. Downloaded from

Clone ID Chr Position (Mb) RP11-192B18 X 86.20 RP11-274M8 X 96.35 RP3-394H4 X 116.43 RP11-481F23 X 134.43 RP1-73A14 X 145.73 RP11-218L14 X 154.32 GS1-225F6 X 154.57 RP11-414C23 Y 2.91 RP11-418M8 Y 8.70 RP11-333E9 Y 12.60 RP11-478I15 Y 17.19 RP11-66M18 Y 22.55 RP11-270H4 Y 26.64

Supplementary Table 1

List of clones represented on the prenatal microarray following the removal of clones as a result of validation experiments. Clone positions are given as midpoints in megabases

http://jmg.bmj.com/

on September 25, 2021 by guest. Protected copyright.