Technology Advances

click to enlarge   Columbus Gallery provides a choice of methods to view, navigate, and analyze high content screening image data—including the ability to view and export intensity profiles from selected regions of interest. (Source: PerkinElmer)

In high content screening (HCS), scientists dig out data with informatics. “Our ability to assay thousands of targets in parallel generates enormous amounts of data, requiring state-of-the-art bioinformatics tools to interpret and integrate the information,” says Berta Strulovici, PhD, vice president, automated biotechnology, at Merck & Co., Whitehouse Station, N.J. She adds, “We have created a system that allows us to easily mine, manage, and share data from high content screening across relevant groups at Merck. This is helping to drive R&D forward more efficiently.” For example, Merck scientists recently found that networks of genes play some role in obesity.

Combining various forms of data could explain basic biology and disease. As Strulovici says, “High content screening will play an important role in connecting our ability to manipulate molecular networks with assessing cellular phenotypes as part of our strategy for target identification and validation. We are essentially taking Merck’s systems biology expertise to the next level by testing these networks and validating new targets.” For instance, Merck can study a gene network by silencing any individual gene or pairs of specific genes. “Coupling this capability with high content, multi-parametric assay methodologies in a robust, parallel way in live cells,” says Strulovici, “is not only a powerful tool for perturbing networks on the cellular level, but it enables us to generate hypotheses at the early pre-clinical stage that can be further validated in the clinic.”

Other companies also drive this field. For example, PerkinElmer just unveiled its Columbus Gallery. “This product offers an ‘open’ link between bioinformatics databases and high-content data,” says Gabriele Gradl, PhD, global product leader for high content screening at PerkinElmer Cellular Technologies, Hamburg, Germany. “The standardized, open-source format for input of images, metadata, and results combined with the standardized format for data output make the Columbus Gallery a powerful tool not only for high content screening but also for many kinds of imaging applications.” This product was also designed to work well with other high technologies, including Intel’s multi-core processor family.

Some advances also come from updates. For example, Accelrys, San Diego, Calif., released a beta version of Pipeline Pilot 7.0. Tim Moran, Accelrys’ imaging product manager explains: “Updates include many valuable tools for high content screening, including plate-data analytics, gene expression, advanced imaging, and data modeling collections.” At the High Content Analysis Conference in Berlin in early March, Accelrys presented new results produced with Pipeline Pilot 7.0. Moran says, “We highlighted the combined use of our data modeling tools in conjunction with our image analysis tools in order to develop a model to predict a compound’s ability to promote apoptosis in differentiating neuronal stem cells using only image fingerprints from Hoechst nuclear stain.”

About the Author
Mike May is a publishing consultant for science and technology based in Minnesota.

This article was published in Drug Discovery & Development magazine: Vol. 11, No. 5, May, 2008, pp. 16-17.

click to enlargeTo make high content screening tools easier to use, febit and InforSense are teaming up to build products that combine advanced algorithms with graphical user interfaces. (Source: febit) HCS Teamwork “In general, genomics tools for microarrays and high content screening have been so successful over the past couple of years that they created data overflow,” says Peer Stähler, chief scientific officer at febit, Heidelberg, Germany. He believes that tools need to return the control of bioinformatics—from building workflows to analyzing results—to research scientists, instead of IT specialists. Consequently, febit collaborated with InforSense, London, UK, to create workflow tools that handle data from various algorithms. InforSense provides the graphical user interface, and febit develops the algorithms. “We are concentrating on making it easy for customers to extract data from microarrays,” says Stähler. “With just a few mouse clicks, they can make a report.” Perhaps most important of all, Stähler says that such tools can be made quickly for new algorithms. “Working with InforSense,” he says, “we can spin out tool after tool for every problem that a microarray user might have.”