Reproducibility, Fidelity, and Discriminant Validity of mRNA Amplification for Microarray Analysis from Primary Hematopoietic Cells

摘要: Analysis of gene expression in clinical samples poses special challenges, including limited RNA availability and poor quality. Quantitative information regarding reliability amplification methodologies applied to primary cells representativeness resulting profiles is limited. We evaluated four protocols for from peripheral blood mononuclear cells. Results obtained with 100 ng or 10 amplified using two rounds cDNA synthesis vitro transcription were compared control 2.5-μg processed a single round transcription. Samples hybridized Affymetrix HG-U133A arrays. Considerable differences results different protocols. The optimal protocol resulted highly reproducible (r = 0.98) good correlation between 0.94). Using the dissimilarities normal individual patient myelodysplastic syndrome primarily maintained after controls. conclude that small variations methodology introduce considerable distortion too strong focus on very number genes picked an array analysis could be unduly influenced by seemingly acceptable methodologies. However, it possible obtain representative optimized