Fluorescence in situ hybridisation (FISH) validation of chr3p26.3 copy number variations (CNVs). (A) Validation of chr3p26.3 microduplication. FISH analysis confirmed the chr3p26.3 microduplication in subject 1 and his father (FISH images from subject 1 pictured). Panel order from left to right: composite image, left flanking clone (yellow), right flanking clone (green), clone within the CNV (red). (B) Validation of chr3p26.3 microdeletion. FISH analysis confirmed the chr3p26.3 microdeletion in 2B, 2C and their father (FISH images from subject 2C pictured). Panel order from left to right: composite image, left flanking clone (yellow), right flanking clone (green), clone within the CNV (red). (C) Normal chromosome 3. (D) Chromosome 3 with duplication. Comparison of red signal intensity in (C) versus (D) demonstrates a duplication. (E) Ideogram of clone locations. Three bacterial artificial chromosomes (BACs) were selected for each CNV: a left clone flanking the CNV pseudocoloured yellow (RP11-204C23), a right clone flanking the CNV pseudocoloured green (RP11-95E11), and clone within the CNV pseudocoloured red (RP11-129K1).

Diagram of polymerase chain reaction (PCR) primer design. (A) Amplification of chr3p26.3 deletion specific product. Primers DelF and DelR were designed to amplify a 454 bp product specific to the observed deletion. In normal genomic DNA, there primers are 670 784 bp apart and do not generate a product. (B) Amplification of chr3p26.3 duplication specific product. Primer DupF and Duper were designed to amplify 520 bp product specific to the observed tandem duplication. In normal genomic DNA, these primers are 790 018 bp apart and face opposite directions.

Fine Mapping of copy number variation (CNV) end points. (A) A NimbleGen fine tiling array further defines the deletion end points. A NimbleGen fine tiling array spanning positions 1 900 000–3 100 000 on chromosome 3 was used to map the deletion in greater detail. The log2 ratio (test/reference) deviates from 0 at positions ∼2 205 400–2 859 400, indicating the deletion breakpoints on chromosome 3. (B) A NimbleGen fine tiling array further defines the duplication end points. The same fine tiling array utilised in (A) was used to map the duplication in greater detail. The log2 ratio (test/reference) deviates from 0 at positions ∼2 003 900–2 795 000, indicating the duplication breakpoints on chromosome 3. (C) Polymerase chain reaction (PCR) across the deletion. PCR with primers spanning the deletion yielded a 454 bp product in subject 2C (A02), subject 2B (A03) and their father (A06). No product was present in subject 2A (A04), their mother (A05), a normal control (A07) or water (A08). See also fig 3A. (D) PCR across the duplication. PCR with primers between the duplicated regions on chr3 yielded a 520 bp product in subject 1 (A02) and his father (A04). No product was amplified from his mother (A03), a normal control (A05) or water(A06). See also fig 3B.

Histogram of Alu sequences near CNTN4. Alu sequences containing 100% identity of 75 bases or more to other Alu sequences within 5 Mb were plotted on the UCSC Genome Browser. The position of each Alu element found to match at least one other Alu, using this criterion, is denoted by a vertical bar. The height of each bar represents the number of “hits”.