By Shane Schick
A Toronto-based life science company is trying to accelerate its ability to analyze proteins so that drugs treatments can be more easily personalized to a patient’s individual needs.
Protana Inc. is
rolling out a workflow system that combines TurboWorx Enterprise, IBM’s DB2 Universal Database, WebSphere Application Server, Tivoli Storage Manager and eServer xSeries hardware. The system will help automate Protana’s work in analyzing protein samples from bacteria, fruit flies, mice or human beings through mass spectrometry and cross-referencing it with biological data in public databases such as the one run by the Blueprint Initiative, among other sources.
Prior to creating the IBM/TurboWorks system, Protana CEO Mark Pearson said some employees were working overnight to transfer data files between servers and applications in its multi-vendor computing environment.
“The kind of throughput we needed to deal with was higher than we could possibility staff,” he said. “The aha moment is when you can relate the identity of a protein with the scientific literature.”
Protana’s hope is that discovering how genes are modified in a particular patients and what proteins are affected could help determine what kind of therapy should be used to treat patients. If a patient displays symptoms of breast cancer, for example, there could be between five and 10 different genes involved, and the right drug may depend on mutations of certain genes.
“We’re dealing here with relatively complex situations where we now know how to analyze – if not understand – gene level and expression,” he said. “We would like to categorize patients. We know that that is possible.”
The combined hardware and software has sped up some of these processes by 90 per cent, Pearson said, from weeks to a matter of minutes.
“The human error has been eliminated and focused staff on the problem of proteomics and not the problems of data management,” he said.
Judy Hanover, an analyst with IDC subsidiary Life Science Insight, said many bioinformatics firms are exploring tools from companies like IBM and SAS to improve their knowledge and content management capabilities.
“For each analysis, there is this cascade of petabytes of data. It gets past the scale of one person being able to handle it,” she said. “They need to be able automate that analysis so they can get down to the individual reaction.”
A potentially greater challenge, Pearson admitted, was ensuring the infrastructure can scale companies like Protana dig deeper into genomic and proteomic information, particularly on the storage side.
“What we would have considered a lot today is miniscule tomorrow,” he said.