Biomax Informatics said last month that it has secured several life science customers for the Viscovery data-mining software, meeting a goal that the bioinformatics company set when it purchased the Viscovery business in 2007.
Gerhard Kranner, Viscovery Software’s founder and managing director, told BioInform this week that the Biomax subsidiary has recently added several life science clients to its core customer base, which has traditionally included firms in telecom, finance, retail, and industry.
Martinsried, Germany-based Biomax acquired Viscovery in 2007 for an undisclosed amount from its previous owner, Vienna, Austria-based Eudaptics Software.
The Viscovery Data Mining Suite provides several modules for predictive analytics and data mining that enable users to explore complex data, recognize relationships, create scoring models, and define target groups. The suite includes Viscovery Profiler, Viscovery Predictor, Viscovery Scheduler, Viscovery Decision Maker, and the Viscovery One(2)One Engine.
At the time of the acquisition, Biomax CEO Klaus Heumann told BioInform that the Viscovery Suite would be marketed as a data-mining extension to its BioXM platform (BI 09/28/2007).
Since the acquisition, Viscovery has aligned the interfaces between both systems and released several versions of its software with extended and improved functionalities, such as statistical group comparison and multi-variate testing corrections, which are useful for gene-expression analyses and text-mining applications, Kranner said in an e-mail.
Kranner explained that Viscovery provides a “backbone for performing explorative data mining, predictive modeling, and analysis of feature vectors resulting from pre-processed experimental and image data.”
Currently, Viscovery is being used in projects within the food production industry and in pharmaceutical and medical research, though the company declined to identify specific customers.
Viscovery's software is also being used in CANCERMOTISYS, a public/private research project funded by the Austrian Federal Ministry of Science and Research and the German Federal Ministry of Education and Research that aims to study the effects of drugs against stomach cancer cell lines. Total funding for the project is €2.4 million ($3.4 million).
Researchers in the project are using the software to explore the relationship between the motility of cancer cells and gene expression data, among other areas.
“The ability of tumor cells to break out of the primary tumor requires increased cell migration …therefore, cellular motility can be used as an indicator for metastasis,” Kranner explained.
“The goal is to predict the type of cellular motility [that would enable] personalized treatment of patients based on the gene expression signature of their cancer cells obtained by a biopsy,” he said.
The company is using a quantitative description of cellular motion to characterize the metastatic behavior of different gastric cancer cell lines. “Additionally, we analyze the gene expression signature and the treatment protocol with an epidermal growth factor receptor inhibitor to determine corresponding influences on the motility,” he said.
The firm faces competition from vendors such as SAS, SPSS/IBM, and Statsoft that also offer data-mining solutions, Kranner said. However, “many of the new approaches to life science problems require the application of a mixture of methods,” he noted, adding that the market is still evolving and there isn’t a “vendor leading in clearly established fields.”
He said that the combined Viscovery and BioXM offering provides a “clear advantage” because it covers “a great part of the requirements in life sciences, starting with the representation and stratification of complex data, through the definition of ontologies and text mining, to the explorative analysis and predictive modeling provided by Viscovery.”
Although he could not name specific life science customers, Kranner explained that the company's software can be used to “optimize processes and product quality in food production” as well as to “identify dependencies between product characteristics — for example, product taste and consumer buying preferences.”
The company said that in toxicology and clinical research, its data-mining capabilities and self-organizing maps can help researchers find potential biomarker candidates by analyzing and clustering clinical and genetic data.
Have topics you'd like to see covered in BioInform? Contact the editor at uthomas [at] genomeweb [.] com.