A certain minimal amount of RNA from biological samples is necessary to perform a microarray experiment with suitable replication. In some cases, the amount of RNA available is insufficient, necessitating RNA amplification prior to target synthesis. However, there is some uncertainty about the reliability of targets that have been generated from amplified RNA, because of nonlinearity and preferential amplification. This current work develops a straightforward strategy to assess the reliability of microarray data obtained from amplified RNA. The tabular method we developed, which utilises a Down-Up-Missing-Below (DUMB) classification scheme, shows that microarrays generated with amplified RNA targets are reliable within constraints. There was an increase in false negatives because of the need for increased filtering. Furthermore, this analysis method is generic and can be broadly applied to evaluate all microarray data.
A copy of the Microsoft® Excel® spreadsheet is available upon request from Edward Bearden.
This is a preview of subscription content, log in to check access.
Kenzelmann M, Klaren R, Hergenhahn M, et al. High-accuracy amplification of nanogram total RNA amounts for gene profiling. Genomics 2004; 83(4): 550–8
Patel OV, Suchyta SP, Sipkovsky SS, et al. Validation and application of a high fidelity mRNA linear amplification procedure for profiling gene expression. Vet Immunol Immunopathol 2005; 105(3-4): 331–42
Li L, Roden J, Shapiro BE, et al. Reproducibility, fidelity, and discriminant validity of mRNA amplification for microarray analysis from primary hematopoietic cells. J Mol Diagn 2005; 7(1): 48–56
Kaposi-Novak P, Lee JS, Mikaelyan A, et al. Oligonucleotide microarray analysis of aminoallyl-labeled cDNA targets from linear RNA amplification. Biotechniques 2004; 37(4): 582–8
Hoffmann K, Firth MJ, Freitas JR, et al. Gene expression levels in small specimens from patients detected using oligonucleotide arrays. Mol Biotechnol 2005; 29(1): 31–8
Lau WK, Chiu SK, Ma JT, et al. Linear amplification of catalyzed reporter deposition technology on nylon membrane microarray. Biotechniques 2002; 33(3): 564, 566-70
Gomes LI, Silva RL, Stolf BS, et al. Comparative analysis of amplified and nonamplified RNA for hybridization in cDNA microarray. Anal Biochem 2003; 321(2): 244–51
Li Y, Li T, Liu S, et al. Systematic comparison of the fidelity of aRNA, mRNA and T-RNA on gene expression profiling using cDNA microarray. J Biotechnol 2004; 107(1): 19–28
Marko NF, Frank B, Quackenbush J, et al. A robust method for the amplification of RNA in the sense orientation. BMC Genomics 2005; 6(1): 27
Kuo WP, Jenssen TK, Butte AJ, et al. Analysis of matched mRNA measurements from two different microarray technologies. Bioinformatics 2002; 18(3): 405–12
Jarvinen AK, Hautaniemi S, Edgren H, et al. Are data from different gene expression microarray platforms comparable? Genomics 2004; 83(6): 1164–8
Rabbani MA, Maruyama K, Abe H, et al. Monitoring expression profiles of rice genes under cold, drought, and high-salinity stresses and abscisic acid application using cDNA microarray and RNA gel-blot analyses. Plant Physiol 2003; 133(4): 1755–67
Ivanov D, Dvoriantchikova G, Pestova A, et al. Microarray analysis of fiber cell maturation in the lens. FEBS Lett 2005; 579(5): 1213–9
Lyne R, Burns G, Mata J, et al. Whole-genome microarrays of fission yeast: characteristics, accuracy, reproducibility, and processing of array data. BMC Genomics 2003; 4(1): 27
Liang M, Briggs AG, Rute E, et al. Quantitative assessment of the importance of dye switching and biological replication in cDNA microarray studies. Physiol Genomics 2003; 14(3): 199–207
Yang IV, Chen E, Hasseman JP, et al. Within the fold: assessing differential expression measures and reproducibility in microarray assays. Genome Biol 2002; 3(11): research0062
Grigoryev DN, Ma SF, Irizarry RA, et al. Orthologous gene-expression profiling in multi-species models: search for candidate genes. Genome Biol 2004; 5(5): R34
Yadetie F, Laegreid A, Bakke I, et al. Liver gene expression in rats in response to the peroxisome proliferator-activated receptor-alpha agonist ciprofibrate. Physiol Genomics 2003; 15(1): 9–19
Vach W. The dependence of Cohen’s kappa on the prevalence does not matter. J Clin Epidemiol 2005; 58(7): 655–61
Cook RJ. Kappa and its dependence on marginal rates. In: Armitage P, Colton T, editors. The encyclopedia of biostatistics. New York: Wiley, 1998: 2166–8
Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome Biol 2002; 3(7): research0034
Uebersax JS. Diversity of decision-making models and the measurement of inter-rater agreement. Psychol Bull 1987; 101: 140–6
Feinstein AR, Cicchetti DV. High agreement but low kappa: I. The problems of two paradoxes. J Clin Epidemiol 1990; 43(6): 543–9
Irizarry RA, Warren D, Spencer F, et al. Multiple-laboratory comparison of microarray platforms. Nat Methods 2005; 2: 345–50
The authors wish to thank Radhakrishnan Nagarajan, Richard Dennis and Eric Siegel from the University of Arkansas for Medical Sciences (UAMS) for their input and support. Also, appreciation is expressed to the National Science Foundation (NSF) Research Coordination Network for facilitating the interchange of ideas in microarray analysis.
This research was supported in part by funds provided to the UAMS Microarray Facility through Act 1, The Arkansas Tobacco Settlement Proceeds Act of 2000, and by NIH grant no. P20 RR-16460 from the Biomedical Research Infrastructure Network (BRIN) Program of the National Center for Research Resources.
Sequences for the primer pairs are given in table AI.
Appendix B: A Comparison of Dissimilar Slides
In order to see how the DUMB table responds when comparing two dissimilar arrays, we compared another pair of dye-swapped slides. However, in this case, we calculated the ratios as channel 1 over channel 2 in both (i.e. the dye-swapped slide was not treated as a dye swap). In this case, the expression levels should be exact opposites of each other. Table AII is the DUMB table for this comparison with channel sum 1000 and fold threshold 1.4 for both arrays. All the parameters of agreement were considerably lower than in previous comparisons. Of particular note are the NDU and both kappa statistics. This example also demonstrated that the channel sum filter and fold-change criteria can be adjusted in a manner that will increase the apparent agreement. Since the dye-swapped slide was not treated properly in this case, higher numbers in the DU and UD cells were expected. Adjusting the channel sum filter eliminated these expected differences. As the channel sum filter was lowered, the differences became more obvious. However, this is not a shortcoming of the table; instead it reinforces the need to examine the parameters of agreement rather than relying solely on the values of the cells.
About this article
Cite this article
Bearden, E.D., Simpson, P.M., Peterson, C.A. et al. Assessing the Reliability of Amplified RNA Used in Microarrays. Appl-Bioinformatics 5, 67–76 (2006). https://doi.org/10.2165/00822942-200605020-00001
- Venn Diagram
- Percent Agreement
- Label Target
- Tabular Method
- Fold Threshold