This work answers the question of whether it is necessary to hybridize individual instead of pooled RNA samples on microarrays for screening gene targets suitable as diagnostic tools for radiation exposure scenarios, while at the same time meeting comparable microarray quality criteria. For developing new clinical diagnostic tools, a two-stage study design was employed in five projects. At first, pooled and not individual RNA samples were hybridized on microarrays for screening purposes. Potential gene candidates were selected based on their fold-change only. This was followed by a validation/quantification step using individual RNA samples and quantitative RT-PCR. Quality criteria from the screening approach with pooled RNA samples were compared with published data from the MicroArray Quality Control (MAQC) consortium that hybridized each reference RNA sample separately and established quality criteria for microarrays. When comparing both approaches, only insignificant differences for quality criteria such as false positives, sensitivity, specificity, and overall agreement were found. However, material, costs, and time were drastically reduced when hybridizing pooled RNA and gene targets applicable for clinical diagnostic purposes could be successfully selected. In search of new diagnostic tools for radiation exposure scenarios, the two stage study design using either pooled or individual RNA samples on microarrays shows comparable quality criteria, but the RNA pooling approach saves unique material, costs, and efforts and successfully selects gene targets that can be used for the desired diagnostic purposes.
Health Phys. 103(2):159Y168; 2012
*Clinic for Hematology, Hemostaseology, Oncology and Stem CellTransplantation, Hannover Medical School, Hannover, Germany; †Department of Nuclear Medicine, Technische Universität München, Munich, Germany; ‡Bundeswehr Institute of Radiobiology affiliated tothe University of Ulm, Munich, Germany; **Department of Urology, Federal Armed Forces Hospital, Hamburg, Germany; §Department of Hematology and Oncology, Federal Armed Forces Hospital, Ulm, Germany.
The authors declare no conflict of interest.
For correspondence contact: Matthias Port, Clinic for Hematology, Hemostaseology, Oncology and Stem Cell Transplantation, Hannover Medical School, Carl Neubergstr. 1, 30625 Hannover, Germany, or email at firstname.lastname@example.org.
(Manuscript accepted 11 January 2012)