The Science Advisory Board - Protocols, Product Reviews, Member Forum, and Science News

Applications and Advancements in DNA Microarray Technology
Over the last decade, DNA microarray technology has evolved to become a powerful tool with which scientists can analyze the expression levels of thousands of genes simultaneously. Because of their small size and the fact that they can contain a large number of genes, DNA microarrays are useful for undertaking genomic-level assessments on a microscopic scale. For instance, DNA microarrays may be used to measure gene expression levels within a single sample or to compare gene expression patterns between two different samples.

Given its potential to significantly advance the understanding of gene expression and function at the cellular level, The Science Advisory Board undertook a study of 500+ researchers who regularly use DNA microarrays in their research. This study specifically focused on ascertaining experimental parameters, throughput levels and supplier preferences. It also measured researchers’ current degree of satisfaction with DNA microarrays and their unmet technology needs.

The majority of scientists who use DNA microarrays do so to determine tissue-specific patterns of gene expression in human samples. The second most popular application is to elucidate patterns characteristic of disease states in human samples. Regardless of their ultimate research goal, however, all scientists must decide between performing their experiments themselves (either with commercial or homemade arrays) or utilizing a core lab. Whatever their preference, most researchers do perform an early step in their DNA microarray experimentæsample preparation-and a late stepædata analysis-in their own labs. Intermediate steps such as sample labeling, hybridizing and reading arrays are typically performed by a core lab perhaps suggesting that some labs do not have the necessary instrumentation (i.e., scanners) to perform these steps themselves.

Of The Science Advisory Board members who participated in this study, 55% utilize solely commercial DNA microarrays, 27% use both commercial and homemade DNA microarrays, and 19% exclusively employ homemade DNA microarrays (made in either their own lab or a core lab). Affymetrix has long dominated the commercial DNA microarray space in part due to their early investment in the technology and supplies 75% of scientists who perform experiments with commercial DNA microarrays. While Affymetrix remains the leading supplier, scientists have begun turning to other suppliers such as Applied Biosystems and Bio-Rad in the past three years.

Because of the high throughput nature of microarrays, a significant amount of data is generated with each experiment. Therefore, having the appropriate software is absolutely essential for collecting and interpreting this information. A significant challenge facing scientists who regularly use DNA microarrays continues to be the ability to integrate with other databases and to compare experimental results across instrument platforms, between different labs or between experiments performed in the same lab.

“We use instrumentation and data software analysis tools from the same provider to eliminate: data compatibility issues, costs, lengthy development steps, defragmentation of know-how and expertise.” – Pharmaceutical Researcher, Austria

Researchers are universally concerned with integrating their results with other databases. This problem has been long-standing in the microarray field, due in part to the “closed” nature of several leading supplier’s platforms. A related complaint among the scientists we polled is their inability to easily share microarray data. The choice of data analysis software supplier also affects data analysis issues. As an example, scientists who use software from Agilent, which utilizes “open platform” type architecture, report having less difficulty normalizing their results. In an effort to surmount some of these most vexing shortcomings, 11% of researchers design custom in-house software. Of the remaining scientists who use commercially developed software, Affymetrix is the primary supplier, with 29% of the researchers selecting them as the company they use most often, followed by 16% who utilize Agilent Technologies products.

“While we use the hardware that has been optimized for the particular microarrays we use (same company), we use a variety of different software packages to allow us to sift as much information from the data as possible. Even when performing the same aspect of the analysis, we will look at the data with more than one software package to evaluate whether different ways of looking at the question will get us the same answers. So there are very tight technical constraints on the performance of the bench science, with a strong need for highly reproducible experimental conditions for purposes of comparing data between chips, requiring that we not using some newer chips under different conditions from the first chips run. But once we have these data from this tightly reproducible set of experimental conditions, we have a lot of freedom to data mine in a variety of ways. In addition, once we have the data sets in hand, if we later decide to start using a new piece of software, we can use it to go back and look at the earlier data sets as well as the newer ones, so use of new software does not threaten issues of reproducibility and comparability across experiments, where a change in hardware would eliminate comparability.” – Principal Investigator, United States

Because DNA microarray experiments generate vast amounts of data, mining for biologically relevant and statistically valid information can be quite challenging. “Surprisingly, there remains little standardization of microarray formats despite the fact that they have been in existence for over 10 years and even are referenced in investigational new drug applications submitted to the Federal Drug Administration (FDA),” asserts Tamara Zemlo, Ph.D., MPH, Executive Director, The Science Advisory Board. According to the results of our study, one of the most significant technical barriers to the growth of DNA microarrays has been the lack of compatibility between data formats, equipment, interfaces and protocols. “Developing, testing and then implementing standards to streamline DNA microarray utilization will increase both the relevance and validity of results obtained from these analyses,” explains Zemlo. Researchers believe that suppliers can best improve DNA microarray technology by developing standardized protocols for all phases of microarray workflow and analysis and improving the signal to noise ratio by a factor of 100.

“With regards to microarrays in general I feel very strongly that there should be more of a consensus regarding the methods of analysis as so many people use so many different methods and these often result in very subjective data.” – Postdoctoral Fellow, United Kingdom

With these improvements to DNA microarray technology, scientists hope to be able to deduce probable functions of new genes by comparing their expression patterns with those of known genes. Such investment in DNA microarray infrastructure promises to allow for a better characterization of gene families, the discovery of new gene expression patterns and the identification of novel genes. Additionally, it will ultimately allow scientists to translate these results to the clinic whereby genes involved in disease causation and progression can be identified.

###
For insights into what life scientists view as the greatest obstacle to standardization of DNA microarrays and their views on bundling of DNA microarray instrumentation and software packages, please click here.

Other articles on DNA microarrays on The Science Advisory Board Web site can be found at:

Anatomy of a Comparative Gene Expression Study

Designing and Enhancing Software: A Bioinformatician's Life

The Tools & Techniques of Drug Discovery: DNA Microarrays

[ View Current & Future Studies ]

[ View Past Studies ]