TaqMan low density array is roughly right for gene expression quantification in colorectal cancer

BackgroundTaqMan low density array (LDA) is promising for high throughput screening in functional genomics by simultaneously measuring mRNA expression of multiple genes. However, the reproducibility and reliability remain to be explored.MethodsWe applied LDA to detect mRNA expression of 95 gastrointestinal differentiation associated genes in 27 colorectal cancers with individual-matched normal mucosa. Conventional Q-PCR assay was done to detect 18 differentially expressed genes in additional 22 colorectal cancers.ResultsA total of 97.2% (11,520/11,844) gene samples were successfully amplified by LDA. There was a perfect agreement between intra-LDA assays in all gene samples (CCC = 0.952, p < 0.0001). Seventy-nine genes showed perfect or substantial agreement between intra-LDA tests (CCC > 0.713). Genes with low Ct values (< 30 cycles) had more genes showing perfect agreement, less showing moderate agreement, and lower ΔCt variances between intra-plate assays than that with high Ct values (> 30 cycles) (p < 0.01). All 18 genes showed the same directional changes in colorectal cancers versus normal mucosa by both SYBR Green and LDA approaches.ConclusionsLDA is a roughly robust method for gene quantification in colorectal cancer, but its reproducibility decreased in low copy genes. Hence, we strongly recommend caution when analyzing LDA results of those low copy genes.