For this reason, we examined the microarray data for comparability between platforms by reviewing liver sample to prostate sample expression values with two different levels: rank correlation of the log-ratio as qualitative assessment, and the microRNA list agreement (detection call and identification of differentially expressed microRNAs) as qualitative assessment.
Consistent with recent studies [ 37- 40], our results show that a thorough standardisation of the array and experiment design, protocols and data analysis procedures, can greatly improve microarray data quality and comparability.
Moreover, the introduction of the Minimum Information About Microarray experiments (MIAME) as standard documentation for array experiments and in transcriptomic databases, increasing the value and comparability of microarray data [ 15].
As regard the technical aspect of this study, it should be added that the comparability of mRNA microarray data (cohorts A and B) with RT PCR data (cohort C) regarding gene expression levels might be a subject of discussion.
Note that in order to achieve comparability between the microarray data from the different sources, all the raw CEL files from this study as well as those downloaded from GEO (Table 1) were imported into Partek Genomics Suite (v6.5) and GC-RMA normalised in the same manner upon import to ensure numeric comparability across arrays.
We excluded 13 datasets contributing 185 TNBC cases from the discovery cohort because they did not fulfill our criteria of comparability of the microarray data (for details see Additional file 4, Supplementary Methods Section 1 and Additional file 1, Supplementary Figure S2).
This transformation serves both to aid the prioritization and to facilitate better comparability between microarray datasets, as it has been suggested that rank-based transformations of microarray data alleviate some of the issues associated with comparing cross-platform, cross-laboratory data (Irizarry et al., 2005).
This experimental design and the resulting datasets provide a unique opportunity to assess the repeatability of gene expression microarray data within a specific site, the reproducibility across multiple sites, and the comparability across multiple platforms [23].
Our approach thus ensures comparability to previously published in vitro data (including previous microarray data, for example [ 26, 36- 38]).
Some initiatives are raising awareness of the effects of variables that might hamper data comparability and are working toward developing best practice guidelines for microarray-based measurements (Hopkins et al. 2004).
