Turned up to eleven: Fair and Balanced

Thursday, June 06, 2002

"High Throughput"

Getting back to science, I think that it is worth approaching this genomics/proteomics thing, slowly, and with an eye toward understanding its promise and limitations. Many people will not be familiar with the words in boldface, but those who have scanned the technical molecular biology literature or written a grant in the past few years will.

Genomics is the study of the entire genome of an organism, or comparative studies on the genomes of several organisms (the genome is the entire set of genes that the organism has). This term can and has been used to describe "sub-genomic" studies, for instance describing a single chromosome of the Human Genome in detail (a comparison with the mouse genome to human chromosome 16 was recently published in Nature), and typically genomics is characterized by the use of a DNA Microarray, often called a "gene chip" (click the link for lots of info on what they are, and how they work). Basically, a gene chip contains sequences corresponding to some or all of the genes in a genome. The sequences are fixed to a solid support, such as a glass slide, chemically, and then specially labelled DNA or RNA is placed on the slide. Because of DNA and RNA complementarity (A=T, G=C), the DNA or RNA from the sample sticks to the spots of fixed DNA, and the output can be read by flourescence imaging (typically shown in publications as red, green, and yellow spots).

This technology is incredibly powerful. Genome sequencing, which is a fairly mature technology (on the timescale of molecular biology), has lead to the exponential growth of new information about the genomes of organisms from viruses to bacteria to rodents to humans. This has allowed for the construction of microarrays of varying specificities, and robotics (an example of a DNA Microarray Robot) allow for rapid, automated production of these arrays. So, what are some of the things that this biotech revolution can do?

1) Gene expression- By purifying mRNA (recalling the "central dogma", DNA to RNA to Protein) from cells under various conditions, we can look at changes in gene expression. Using animal models, we can also look at gene expression in vivo.

2) Massively parallel experiments- Designing an experiment with a "chip" allows for the experimenter to find out about genes or gene expression all at once, rather than looking at one or a handful of genes at a time. This is the "high-throughput" in the title (its also one of my favorite made-up words).

3) Complex gene expression patterns and "pleiotropic" regulation of genes. A severe limitation of "old-school" genetics is that it is difficult to look at more than a few genes at a time, and it therefore took (and still takes) a long time to figure out what regulatory protein is controlling expression of what other genes, and the particulars of when and how this occurs. DNA microarrays are a powerful tool (but by no means the only tool) for uncovering these relationships faster.

As time goes by, many more applications of this technology will be realized (there are probably others right now that I am not aware of). Of course, I have listed general applications of the technology, not specific research goals. Needless to say, using this approach to drug discovery and drug evaluation has been a cornerstone of DNA microarray development (the old profit motive strikes again!).
I am going to leave it here for a little while, and come back with some thoughts on the limitations of genomics in general, and microarray technology specifically.