Gene Expression/RNA Biomarker Analysis

Biopharma laboratories process a large volume of data from information management systems, which is why bioinformatic tools and concordance of data across technology platforms are essential components of the biomarker discovery and analytical validation process, ensuring that all systems are connected, accessible, and reliable.

Applied Biosystems solutions enable comprehensive, high-quality, reproducible, and scalable gene- and transcriptome-level analysis across platforms using integrated data tools to help identify accurate biomarker signatures to translate discovery into potential future clinical utility.

RNA sequencing (RNA-Seq), gene expression microarrays, and qPCR provide comprehensive solutions for gene expression analysis and biomarker discovery, screening/profiling, and verification. The different RNA analysis platforms measure absolute quantities of transcripts by using very different methods and algorithms, making comparisons of absolute expression levels complicated. However, concordance of data is essential so that results can be compared across platforms and without having to repeat experiments, minimizing error and freeing up time and instruments in the lab. Since qPCR produces relative gene expression measurements, comparing gene expression differences between samples with qPCR is the most relevant approach for benchmarking RNA-Seq and gene expression microarrays.

Applied Biosystems TaqMan Gene Expression assays have long been considered the gold standard for studying gene expression, providing a wide dynamic range of greater than six logarithmic units of expression levels, high sensitivity, and high specificity. Figures 1 and 2 below show concordance of data generated using TaqMan Assays with the Ion AmpliSeq Transcriptome Human Gene Expression Kit and Applied Biosystems Clariom D Assay results for differentially expressed genes in two reference RNA samples. The same set of 70 genes was investigated across all three platforms. Note the high degree of concordance between the two comparisons based on differential expression status (i.e., differentially expressed or not differentially expressed).

Figure 1. Fold-change correlation between the Ion AmpliSeq Transcriptome Human Gene Expression Kit and TaqMan Gene Expression assays. The scatter plot shows that the data sets are highly correlated. Ion AmpliSeq data were analyzed on Torrent Suite Software, and qPCR data were analyzed on Connect, the cloud-based platform from Thermo Fisher Scientific, using the RQ app.

Figure 2. Fold-change correlation between a Clariom D Assay and TaqMan Gene Expression assays. The scatter plot shows that the data sets are highly correlated. Clariom D Assay data were analyzed on TAC 4.0, and qPCR data were analyzed on the Connect platform using the RQ app.

Expression quantitative trait loci (eQTLs) are genetic variants associated with changes in gene expression that have become increasingly important in connecting genome-wide associated studies (GWAS) results to molecular mechanisms of various diseases. eQTLs are identified by linking variations in transcript abundance with variations in genotypes (Figure 3A). Typically, this involves collecting transcriptomic and genomic variant data from samples in a large cohort, and correlating the difference in gene mRNA levels with SNP genotypes in individuals.

eQTLs can fall into two categories. cis-eQTLs are SNPs that are proximal to a target locus (Figure 3B). Proximity is variably defined, but typically cis-eQTLs fall within 200 kb–1 Mb of the transcriptional start site of a gene. Thus, these are usually thought of as SNPs that fall within the promotor, enhancer, or other transcriptional regulatory sequences. The other type of eQTL, trans-eQTLs, are usually genetically unlinked and are thought to regulate expression of a locus from a distance. For example, a trans-eQTL could be a SNP in a transcriptional activating protein that lowers, but does not eliminate, the DNA-binding affinity of the protein to all binding sites (Figure 3C). eQTLs can also be tissue specific—a SNP that affects the expression of a gene in one tissue may have no effect in a different tissue.

Figure 3. How eQTLs affect expression levels of genes. (A) In a wild type individual, a transcription factor (TF) binds to the controlling region of two genes, resulting in normal transcript levels. (B) An individual with a cis-eQTL can have a SNP in the region bound by the transcription factor that reduces (but may not eliminate) transcript levels of gene B while leaving levels of gene A unaffected. (C) In an individual with a trans-eQTL, a SNP in the DNA-binding motif of the transcription factor reduces (but may not eliminate) its affinity to all binding sites, resulting in reduced expression of multiple genes.

A link between the GWAS-identified SNP locus and eQTL suggests that differential expression of the gene affected by the eQTL could contribute to the trait. Once a putative link has been established, verification of the link is needed. This could be done by research screening the transcriptome and genome of even more samples; however, this could become unwieldy and expensive. Alternatively, an orthogonal technology, such as qPCR, could be used to confirm the association of the SNP genotype and mRNA levels of the affected gene in a targeted manner. This method has the potential to deliver information on targeted associations in a cost-effective approach.

The right tools for the job: find your ideal match

Which gene expression technologies are right for you?

Featured gene expression technologies

• Microarray instruments for whole transcriptome analysis
• Ion GeneStudio S5 series of instruments for targeted sequencing
• QuantStudio 3D Digital PCR System for rare mutation analysis
• SeqStudio Genetic Analyzer for Sanger sequencing of single genes
• QuantStudio real-time PCR instruments for discovery and validation

Related applications

• Cancer research
• Microbial genomics
• Drug discovery and development
• Stem cell research