This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Coe, B. P. Articles by Meijer, G. A. Search for Related Content PubMed PubMed Citation Articles by Coe, B. P. Articles by Meijer, G. A. Related Collections Related Article

Comment re: A Comparison of DNA Copy Number Profiling Platforms

British Columbia Cancer Research Centre, Vancouver, British Columbia, Canada

Bauke Ylstra, Beatriz Carvalho and Gerrit A. Meijer

VU University Medical Center, Amsterdam, the Netherlands

To the Editor:

We read with interest the publication by Greshock and colleagues (1). Side by side comparison of array comparative genomic hybridization platforms has become a hot topic and several reports have aimed to evaluate various arrays to facilitate the user's selection of the appropriate platform (2–5). Greshock and colleagues used an interesting and meaningful approach to compare the sensitivity and specificity of three popular oligonucleotide comparative genomic hybridization platforms by receiver operating characteristic curve analysis of actual experimental results. The finding that the averaging of multiple array elements is platform-dependent and is required to reduce noise on some platforms matches well with previous reports (2, 3).

Previous reports disagree with respect to two pivotal issues, resolution and coverage. The usage and definition of the term "resolution" in the manuscript is debatable. Greshock and colleagues define resolution in terms of mean probe spacing, which distorts the estimate of true performance when array elements are unevenly distributed. Thus, the genomic distribution of array elements has to be considered in determining resolution. For example, the Affymetrix 500K (AF500K) array is defined in the manuscript as having a resolution of 17.7 Kbp (1), whereas distribution-dependent metrics have defined the array's performance at this level as allowing detection of 58% (2) of 1.5-fold alterations this size. In fact, our study suggests that AF500K is only likely to reliably (95%) detect alterations of 75 Kbp (2).

Additionally, the authors state that the Affymetrix AF500K array, offers "broader and more even coverage across the genome" than the Agilent AG244K array (1). However, we found similar detection probabilities between these platforms at all alteration sizes (2). Our finding contradicts the authors' statement as 244,000 elements should not detect alterations at the same frequency as 500,000 elements, unless the AF500K platform is less evenly distributed. These conflicting results can perhaps be explained by the comparison of only threshold-segmented regions of high-level alteration. Given the biases induced in probe reduction, including restriction digestion combined with PCR (AF500K) as opposed to Random Prime–based labeling (AG244K) it is likely that specific ratio cutoffs may detect different regions. For this reason, the use of automated segmentation algorithms for the comparison of detected regions as shown by Hehir-Kwa and colleagues (3), Paris and colleagues (4), and Wicker and colleagues (5) is likely a more accurate methodology.

Overall, we appreciate the importance of this study; however, we believe that the discussion of resolution and comparison of alteration detection is inaccurate.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.

References

1. Greshock J, Feng B, Nogueira C, et al. A comparison of DNA copy number profiling platforms. Cancer Res 2007;67:10173–80.[Abstract/Free Full Text]
2. Coe BP, Ylstra B, Carvalho B, et al. Resolving the resolution of array CGH. Genomics 2007;89:647–53.[CrossRef][Medline]
3. Hehir-Kwa JY, Egmont-Petersen M, Janssen IM, Smeets D, van Kessel AG, Veltman JA. Genome-wide copy number profiling on high-density bacterial artificial chromosomes, single-nucleotide polymorphisms, and oligonucleotide microarrays: a platform comparison based on statistical power analysis. DNA Res 2007;14:1–11.[Abstract/Free Full Text]
4. Paris PL, Sridharan S, Scheffer A, Tsalenko A, Bruhn L, Collins C. High resolution oligonucleotide CGH using DNA from archived prostate tissue. Prostate 2007;67:1447–55.[CrossRef][Medline]
5. Wicker N, Carles A, Mills IG, et al. A new look towards BAC-based array CGH through a comprehensive comparison with oligo-based array CGH. BMC Genomics 2007;8:84.[CrossRef][Medline]

This Article Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Coe, B. P. Articles by Meijer, G. A. Search for Related Content PubMed PubMed Citation Articles by Coe, B. P. Articles by Meijer, G. A. Related Collections Related Article