Practical uses of quality assessment for high-dimensional gene expression data

Abstract

A new generation of high-throughput measurement technologies in genomics has opened up new avenues for biomedical research. To make use of this potential, statistical challenges related to the size and complexity of these new types of data sets need to be overcome. In particular, noisy data has led to irreproducible scientific results undermining the credibility of the new technologies. This article reviews recent work by statisticians on visualisation and assessment of the quality of data from gene expression microarrays and related technologies. It then traces the impact this work had on the biomedical research community. An example of the use of the new statistical quality assessment tools is their role in the Microarray Quality Control project, an FDA initiative to establish quality standards for high-throughput gene expression data. Another example is their role in the development of a diagnostic tool for thyroid cancer that has hugely reduced the number of unnecessary surgeries.