Microarray Centre

Intelligent Life at MaRS

By Nancy Eaton

Carl Virtanen, manager of bioinformatics at the UHN Microarray Facility, examines a microarray slide from a specialized laser scanner designed to read genetic information found on the slide surface. Virtanen’s department annotates this data and includes it in web-based databases used by biomedical researchers worldwide.

When Carl Virtanen became the manager of bioinformatics at the Microarray Core Facility at Toronto’s University Health Network (UHN), he learned that running a successful department could also have a downside. “We were growing so fast that we were quickly running out of room,” he says. “I was always looking towards the future of having a dedicated room to hold our clusters and all our equipment.”

Virtanen got his wish when he moved to MaRS.

No, he didn’t move to outer space. He and the Microarray Core Facility moved to an exciting new inner space — the premier spot in a state-of-the-art facility called the MaRS Centre. Founded and developed to foster and incubate Canada’s biotechnology, bioinformatics, and nanotechnology industries, the MaRS Centre wants to make it easier for Canadian companies and organizations to fast-track biomedical products and technologies.

Managing Mountains of Data

In moving to the Toronto Medical Discovery Tower at MaRS in 2005, Virtanen seized the opportunity to overhaul the UHN Microarray Core Facility’s aging IT infrastructure to better support its goals. And that included switching to Mac OS X, Apple Xserve, and Xserve RAID as the key solution for efficiently managing their massive amounts of data, while minimizing overhead. Apple Xserve and Mac OS X allowed Virtanen and his staff to focus more on their scientific mission instead of attending to time-consuming IT issues — a mission which includes generating, reporting, and analyzing data of critical importance to researchers worldwide.

Virtanen’s department produces microarrays — slides with tens of thousands of bits of genetic information on them — for UHN clients Princess Margaret Hospital, Toronto General Research Institute, and Toronto Western Research Institute, as well as other provincial, national, and international clients. In addition, they function as service providers for commercial microarray vendors, including Agilent and Affymetrix.

Researchers use the information gleaned from the microarrays to compare healthy and diseased tissue at the genetic level, with the goal of determining where and how things go awry biologically. They use this information in any of several ways — in diagnosing medical conditions, in drug discovery, and for pure research.

Annotated Databases Accessible to All

Each microarray slide produced at the UHN Microarray Core Facility includes as many as 35,000 genes, all of which Virtanen’s department annotates and adds to a series of web-based databases, freely accessible to anyone around the world. “When we mine the slides for information, the only way we can interpret what kinds of patterns we see is to have some kind of high-level annotations for them,” Virtanen explains. “We have a database that, at last count, has more than 700 million records stored on our Xserve RAID system that are basically the annotations for what’s on our arrays.”

“We chose Apple equipment because it offered the greatest value per dollar in terms of computing power, storage, and simplicity of administration and setup.”

To provide greater context and make their data even more useful, Virtanen’s department downloads and integrates information four times a year from public databases specializing in particular types of genetic annotation, including PubMed, NCBI, and UCSC. “We bring those databases in-house,” Virtanen says. “And then we run a bunch of annotation algorithms we’ve made to cook them together into a big melting pot, and distill out the most crucial information. All that’s done automatically.”

In addition to producing and annotating microarrays, Virtanen and his team are actively involved in collaborating with researchers to analyze the mRNA expression and epigenetic profiles of major diseases, including cancer and schizophrenia. “Our group is involved at every step, right from the initial design of experiments to the statistical analysis of the end data,” he says.

1 2 Next Page >