Mark Manary and Sean Diehl - Overcoming the Clinical Sample Bottleneck for Expression Studies

Extracting robust expression data from clinical samples represents a unique opportunity to find actionable biomarkers. But it does not come without challenges. Clinical samples such as whole blood and feces are technically challenging to work with, often of poor quality, and provide limited amounts of high-quality RNA making it difficult to generate reproducible and informative expression data.

Feces, in particular, are increasingly identified as useful non-invasive samples for molecular diagnosis of infectious disease, but their composition consists of a complex microbial community making the identification of human RNA signatures problematic. Whole blood is also a desirable sample type as it is widely available and collection is relatively non-invasive and inexpensive. Yet 70% of mRNA in whole blood is globin mRNA released from red blood cells, introducing noise and decreasing the sensitivity for detecting relevant transcripts. Most expression technologies require the removal of globin mRNA from the sample before processing, a time-consuming and costly step that introduces bias. Due to such challenges, researchers are demanding better expression methods for use with clinical samples, like whole blood and feces.

Choosing the right transcriptome profiling tool compatible with clinical samples is key to generating informative biomarkers that can be used to improve human health.