Quantitative or qualitative transcriptional diagnostic signatures? A case study for colorectal cancer

Abstract

Background: Due to experimental batch effects, the application of a quantitative transcriptional signature for disease diagnoses commonly requires inter-sample data normalization, which would be hardly applicable under common clinical settings. Many cancers might have qualitative differences with the non-cancer states in the gene expression pattern. Therefore, it is reasonable to explore the power of qualitative diagnostic signatures which are robust against experimental batch effects and other random factors.

Results: Firstly, using data of technical replicate samples from the MicroArray Quality Control (MAQC) project, we demonstrated that the low-throughput PCR-based technologies also exist large measurement variations for gene expression even when the samples were measured in the same test site. Then, we demonstrated the critical limitation of low stability for classifiers based on quantitative transcriptional signatures in applications to individual samples through a case study using a support vector machine and a naïve Bayesian classifier to discriminate colorectal cancer tissues from normal tissues. To address this problem, we identified a signature consisting of three gene pairs for discriminating colorectal cancer tissues from non-cancer (normal and inflammatory bowel disease) tissues based on within-sample relative expression orderings (REOs) of these gene pairs. The signature was well verified using 22 independent datasets measured by different microarray and RNA_seq platforms, obviating the need of inter-sample data normalization.

Conclusions: Subtle quantitative information of gene expression measurements tends to be unstable under current technical conditions, which will introduce uncertainty to clinical applications of the quantitative transcriptional diagnostic signatures. For diagnosis of disease states with qualitative transcriptional characteristics, the qualitative REO-based signatures could be robustly applied to individual samples measured by different platforms.

Keywords: Batch effects; Classifiers; Diagnostic signature; Platform; Relative expression orderings.

Fig. 1

Quantitative measurement variation for replicates measured by PCR-based technologies. For each of the sample types (sample A and sample B) measured by StaRT-PCR™ Assays and TaqMan® Assays, the red bar denotes the percentage of genes that shows at least 10% CV and the green bar denotes the percentage of genes that shows at least 15% CV. The total number of such genes within each assay and sample type is noted by blue dots connected by lines and is read on the secondary axis

Fig. 2

Sensitivity and specificity of SVM classifiers (a) and naïve Bayesian classifier (b) for validation datasets. Notably, some datasets included only colorectal cancer tissue samples or normal tissue samples, so only the results of sensitivity or specificity were shown for those datasets

Fig. 3

Analysis procedure for identifying a cross-platform REO-based signature

Fig. 4

Performance of k-gene pairs REO-based signature applied to the training set. The majority vote rule was used for classification

Fig. 5

Performance of the REO-based signature applied to multiple independent datasets from different platforms. The majority vote rule was used for classification

Fig. 6

The distribution of the expression levels of the 3 gene-pairs in GSE8671. The gene expression levels of GPAT3 and TRIP13 (a), PYY and CKAP2 (b) and SDCBP2 and DAP3 (c)