Across array comparative genomic hybridization

Aim of this study is to test if a reference channel of a different array can be used to make array CGH more sensitive, cost effective and less labor intensive. The BT474 breast cancer cell line was compared to a mix of normal reference DNAs hybridized on different arrays and days and DNA copy number profiles were evaluated. Quality was assessed, using regular dual channel array CGH as a standard, using four quality measures; i.e. the MAD (median absolute deviation) value of chromosome 2, an amplitude of the ErbB2 gene amplification, a deletion on chromosome 9 and a deflection on chromosome 8. In addition, the use of a reference channel of a different array was tested for genomic DNA derived from formalin-fixed paraffin-embedded tumor tissue samples. This so called across array CGH (aaCGH) analysis yielded slightly higher quality chromosomal copy number profiles when using the same dye compared to analysis using two different dyes, both when exchanging fluorescent signals of hybridizations performed on different arrays and on different days. This approach avoids redundant hybridizations of the same reference material in every experiment and allows hybridization of two test samples on one array. AaCGH analysis substantially reduces cost of consumables and labor while yielding at least equivalent quality as the dual channel procedure. Keywords: array CGH data, across array CGH analysis Genomic DNA from the cell line BT474, two colorectal cancer (CRC) tissue samples (FFPE) and one gastric cancer (GC) tissue sample (FFPE) was isolated. DNA isolated from blood obtained from eighteen healthy males was mixed for the use as a normal reference. Hybridizations were performed on 4x44K arrays (agilent technologies), on different days. Hybridizations of BT474 versus normal reference were performed in triplicate, and also the dyeswap was performed in triplicate. In total, 24 single channel array CGH data was used to calculate log2 tumor to normal ratios