|
Protocols for the assurance of microarray data quality and process control. Burgoon, L.D.1,3,4, Eckel-Passow, J.E.5, Gennings, C.6, Boverhof, D.R.2,3,4, Burt, J.W.2,3,4, Fong, C.J.2,3,4, Zacharewski, T.R.2,3,4*. 1. Department of Pharmacology & Toxicology, 2. Department of Biochemistry & Molecular Biology, 3. National Food Safety & Toxicology Center, 4. Center for Integrative Toxicology, Michigan State University, East Lansing, MI, USA 48824 5. Department of Health Sciences Research, Mayo Clinic Cancer Center, Rochester, MN, USA 55905 6. Department of Biostatistics, Virginia Commonwealth University, Richmond, VA, USA 23298. Microarrays represent a powerful technology that provides the ability to simultaneously measure the expression of thousands of genes. However, it is a multi-step process with numerous potential sources of variation that can compromise data analysis and interpretation if left uncontrolled, necessitating the development of quality control protocols to ensure assay consistency and high data quality. In response to emerging standards, such as the Minimum Information About a Microarray Experiment (MIAME) standard, tools are required to ascertain the quality and reproducibility of results within and across studies. To this end, a comprehensive quality control protocol for microarrays was developed using cDNA microarrays from in vivo and in vitro dose-response and time-course studies. The protocol combines: 1) diagnostic plots monitoring the degree of feature saturation, global feature and background intensities, and feature misalignments with 2) plots monitoring the intensity distributions within arrays with 3) a support vector machine (SVM) model. This work has been supported by NIH grants ES11271, ES12245 and Superfund P42 ES04911. Support for LDB and JEE was provided by T32 ES07255 and NCI grant R25 CA92049, respectively. TRZ is partially supported by the Michigan Agriculture Experiment Station. |
