Historical Perspective

Before entering the mechanics of genomic analysis we will consider three developments that ushered in the current technology. The first event was the refinement of practical genomic assays. These assays are practical in that they are relatively inexpensive, rapid, and widely available. Perhaps the most important was first described in Patrick Brown's lab in the journal Science in 1995 (7). Brown and colleagues developed a method in which single-stranded DNA of known sequence could be mechanically spotted onto a glass slide in a grid pattern. Having made the archetype of what is now known as a DNA microarray (or DNA chip), the researchers isolated messenger RNA (mRNA) from samples under a variety of conditions. The mRNA was reverse-transcribed to complementary DNA (cDNA) and fluorescently labeled before being hybridized against the cDNA spotted on their microarry. The resultant hybridized arrays were quantitatively scanned for fluorescence, producing an image in which the intensity of each spot corresponded to the degree of hybridization present for that cDNA clone. Accordingly the intensity of individual spots was proportional to the mRNA present in the initial sample, confirmed by Northern blot. Although refinements have been made since, this method remains the model for the most widely used assay in the field of genomics.

The second defining event in the evolution of functional genomics was a dramatic improvement in the quality of genomics databases. These improvements occurred as a result of the completion of the Human Genome Project, along with the efforts to systematically name genes and curate gene libraries such as Unigene managed by the National Center for Biotechnology Information (8-11). The patterns generated by DNA microarray experiments have potential uses independent of a full understanding of the genes that comprise the arrays. For example, expression sequence tags (ESTs) of known sequence but unknown function are routinely reproducibly associated with specific disease states (12). However, to make full use of the genomic data, researchers and clinicians need to be able to communicate efficiently using such information as sequence and gene function.

The third component in the genesis of functional genomics relates to the technical improvements in microcomputing, imaging, and statistical methodologies. Just as the microcomputer popularized multivariate regression techniques in the 1980s, the 1990s witnessed progress in computing speed and digital imaging necessary to process genomic experiments. Concurrently, biostatisticians, computer programmers, and computational biologist adapted analytical techniques. Finally, dramatic improvements in Internet technologies allowed both data and methods to be shared efficiently.

Each of the three necessary developments continues to evolve. The DNA microarray itself is constantly being refined in tandem with a wide variety of other genomic assays. Similarly, the initial phase of the Human Genome Project has been completed, but the range of normal human DNA variation continues to be defined. Gene annotation databases, such as the Gene Ontology Consortium's, are updated daily as details on function emerge. Finally, the computational standards of genetic analysis are in a constant state of development. In spite of these ongoing developments, a core knowledge base has emerged and will serve as the focus of this chapter.

Although an initial flurry of genomics activity focused on cell culture, proof of concept, methods of analysis, and simple organisms, publications using human samples emerged relatively quickly (13). By 1999, a significant number of clinical samples were being analyzed using array technologies, and by 2000, we saw the first publications in the Abridged Index Medicus journals (a subset of journals devoted to clinical research). An understanding of basic genomics techniques is vital for those who will soon be called on to evaluate the growing number of clinical applications.

The push toward genomics research underscored by a promise that previously insolvent clinical research problems might now have new solutions. Perhaps the most common difficulty is recruitment of patients in sufficient numbers. Recruitment is a problem for rare disease, but even studies of common disease fail to generate cohorts of sufficient size. When the clinical problems are complex or the biology of the system incompletely understood, the theoretical numbers of patients needed could easily exceed all reasonable recruitment expectations. Significant time and effort investment are required to answer a single clinical question, bearing little fruit when recruitment is not met. The losses can be mitigated by testing multiple hypotheses in one study, although the methodology of probabilistic hypothesis testing frowns on this practice.

As we will see in the following sections, clinical genomics offers an alternative to the limitations of clinical research by two mechanisms. First, genomics offers tools for analyzing studies with small numbers of samples, each sample being associated with many variables. Second, genomics analysis offers the potential to clarify complex biologic systems, including identifying disease subgroups. When subgroups are identified with respect to disease behavior, clinical trials could be designed to target those with the largest expected response to therapy. This process could drastically decrease the required sample size needed for such a study.

The problems of patient recruitment, rare diseases, and complex biology are particularly evident in the field of oncology. The ability to identify tumor subtypes that have characteristic biology or clinical behavior would be a major contribution in terms of reduction of disease complexity. There is good evidence that microarray data might refine tumor diagnosis in cases where light microscopy and special stains are indetermi-nant. Investigators have differentiated carcinomas of unknown primary, primary lung tumors from metastatic colon cancer, and estrogen receptor-positive and -negative tumors (12,14,15).

In addition to identifying know tumors types, previously unrecognized tumor subtypes might be described by clinical outcome, such in the work of Alizadeh in lymphoma or by differences in underlying cancer biology (16). Perhaps for these practical reasons, the promise of genomics techniques in oncology is being embraced the most enthusiastically. An informal Medline review suggests that as many as half of all genomics publications relate to the field of cancer and we will draw on cancer for the clinical examples in this text.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment