부산대학교 도서관

To link to full-text access for this article, visit this link: http://dx.doi.org/10.1016/j.canlet.2005.08.007 Byline: Torstein Tengs (a)(b), Jeff C. Lee (a)(b), J. Guillermo Paez (a)(c), Xiaojun Zhao (a), Thomas LaFramboise (a)(b), Georgia Giannoukos (b), Roman K. Thomas (a)(b) Abstract: We have designed resequencing microarrays to test the performance of this platform when interrogating a large number of exons (164 total) from genes associated with cancer. To evaluate false positive and negative rates, dideoxy sequencing was done for 335,420 bases interrogated by the arrays. From the array data, calls could be made for [approximately equal to]97.5% of the bases, and false positive rates were very low with only a single mutation reported from the array dataset for which the corresponding dideoxy trace had a clean wildtype sequence. For the nucleotide positions where array calls were made, false negative rates were 1.41% for heterozygous mutations. All the homozygous mutations were detected, but 8.11% were erroneously reported as heterozygous changes from the reference sequence by the array analysis software. In addition, 20 non-small cell lung cancer (NSCLC) samples were analyzed using the arrays, and both somatic and germline mutations were found. The most interesting findings were two MET mutations that have recently been implemented in NSCLC. Large scale MALDI-TOF genotyping indicated that one of these mutations (T1010I) might represent a true cancer-causing genotype, whereas the other (N375S) appears to be a common germline polymorphism. Author Affiliation: (a) Department of Medical Oncology, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA 02115, USA (b) The Broad Institute of MIT and Harvard, Cambridge, MA 02139, USA (c) Merck & Co., Inc., 33 Avenue Louis Pasteur, Boston, MA 02115, USA. Article History: Received 9 July 2005; Revised 26 July 2005; Accepted 3 August 2005