NIST Special Publication 1200-23: Preparation of a Set of Total RNA Benchmarking Samples for Performance Assessment of Genome-scale Differential Gene Expression Version 1.0
A variety of technologies have been developed for genome-scale profiling of differential gene expression. The level of gene expression difference is typically reported in the form of a ratio, comparing one sample (or class of samples) to another. Studies to develop cancer biomarkers typically involve comparing tumor tissue to normal tissue and identifying features that distinguish the two classes. However, most published cancer biomarkers never make it into clinical practice , with a major technical challenge being the irreproducibility of results . A set of papers describes assessing technical performance using sets of mixed total RNA samples designed with between-sample differences in the relative abundance of mRNA [3-7]. Recent work within the Early Detection Research Network (EDRN) of the National Cancer Institute  has used these mixed sample sets for measurement assurance of genome-scale microRNA (miRNA) biomarker discovery. This protocol presents a general procedure for the preparation of these mixture sets, for any biomolecule cohort (e.g. mRNA, proteins, etc.), provided mixture components are available that express different abundances of the molecules.
The full publication is available here.