Development of single-stranded array platform for exon array CGH. Schematic diagram shows the approach adopted to make single-stranded PCR products with the increased sensitivity on the array platform described in this article. A, Oligonucleotide containing a 5′-C6-aminolink modification (gray ball) used as a primer (short green bar) on a genomic DNA or PCR template strand (red wavy line) in a PCR reaction. B and C, Synthesized double-stranded PCR product containing the 5′-aminolink modification on one strand only. Newly synthesized strand containing the 5′-aminolink modification is shown as a green dotted/solid line. D, PCR product spotted onto the microarray slide. E, Covalent attachment to the surface formed through the 5′-aminolink modification by subjecting the slide to high humidity. F, Removal of reverse strand that is not covalently attached to the surface of the slide by use of physical and chemical denaturation. G, The resultant single-stranded array element. H, Microarray image (gray scale) showing the results of fluorescently labeled genomic DNA hybridized to test arrays containing single-stranded and double-stranded PCR products for a series of genomic array elements (“i–xii”) spotted in duplicate. Single-stranded elements for either one strand (5′-aminolink on forward strand [lanes 1 and 2]) or the other strand (5′-aminolink on the reverse strand [lanes 5 and 6]) or for double-stranded elements (5′-aminolinks on both strands [lanes 3 and 4]) were prepared on the test array. Single-stranded elements gave a higher signal for all genomic products (see the “Results” section).

Exon copy-number changes in medically relevant disease genes. Representative results of exon array CGH for patient DNA samples hybridized against a female pool that demonstrates the variety of copy-number changes detected in the present study. All patient samples were fluorescently labeled using Cy5, and all control female-pool samples were fluorescently labeled using Cy3. The features of the plots are as described for figure 2. The gray arrows highlight the copy-number changes. A, Female patient—EA02(260055)—with HNPP showing a deletion of the entire PMP22 gene. B, Male patient—EA07(221358)—with CMT1 showing a duplication of the entire PMP22 gene. C, Male patient (DH/PMD2–1) with PMD showing a triplication of the entire PLP1 gene. D, Female PMD carrier (PMD2–2, mother of patient in panel C) showing a triplication of the entire PLP1 gene. E, Male patient (EA19) with DMD showing a duplication of exon 2 in the DMD gene. F, Male patient (p130) with NF2 showing a deletion of exons 13–15 of the NF2 gene. G, Male patient (EA14) with DMD showing a deletion of exons 3–7 of the DMD gene. Panel also shows the confirmatory PCR results; lanes are numbered according to the exons assayed and size markers (“M”), in base pairs, shown at the left of the gel image. H, Male patient (EA12) with DMD showing the deletion of exon 51 of the DMD gene. Confirmatory PCR is also shown with lanes numbered according to the exons assayed and size markers (“M”), in base pairs, shown at the left.

Validation experiments for exon arrays. A, Fluorescently labeled DNA derived from genomic DNA from a single male (Cy5) hybridized on the exon array with that derived from five females (used as a pool) (Cy3). The plot shows the Cy5 channel:Cy3 channel ratios obtained for exons (represented as black dots) for COL4A5, DMD, NF2, PLP1, and PMP22. Each exon is plotted as a function of its position 5′→3′ for each gene. The number of exon array elements assayed for each gene is shown in parentheses below the gene name. Exons for the three X-linked genes and the two autosomal genes showed ratios centered around the theoretical values of 0.5 and 1, respectively. B, Fluorescently labeled DNA derived from genomic DNA from a single female (Cy5) hybridized on the exon array with that derived from the female pool (Cy3). The plot shows the ratios obtained for each exon, as described above (A), centered around the theoretical value of 1. C, The effect of signal:noise ratio on measurement error determined, using exon arrays, in 10 male versus female-pool validation experiments. The plot shows the relationship between mean signal:noise and mean SD. Both means were derived as the mean of all mean quadruplicate element measurements per experiment. The black triangles represent each of the 10 male/female validation experiments. The correlation coefficient (r) between mean SD and mean signal:noise was −0.97.