Spain's Biotech Revolution

Searching for Data

Many researchers who have spent part of their careers in biotechnology with either Genetrix or PharmaMar have moved on to form companies of their own. Juan Carlos del Castillo, formerly with Genetrix, now heads a family of companies called the Bionostra Group. The group’s subsidiaries are involved in a variety of aspects of biotechnology; Chimera Pharma, for instance, focuses on virus-like particles that can be used to develop a variety of vaccines and also serve as the basis for therapeutics.

But the product that elicits the most enthusiasm from del Castillo is a recent release from the bioinformatics company Bioalma: he calls it a Google for printed biological data. The engineers at Bioalma created a system of information retrieval that can search texts and understand the biological meaning of the written word.

“It took seven years of research, and now we have a product that’s out in the world and working very, very well,” del Castillo says proudly.

Thousands of papers are published every day, and it’s nearly impossible for researchers to keep up with the volume. Company engineers needed to develop algorithms for a computer to recognize all the ways of referring to genes, proteins, diseases, symptoms, and other related biological terms, and then create a searchable database of them.

Designing the system proved quite difficult. Says del Castillo, “The challenge for us was to design a computer system capable of reading those papers, understanding what is written, and offering the information to scientists in a structured, comprehensive format.” The end result was software called Alma Knowledge Server (AKS). In addition to purchases of AKS by companies such as Roche and Novartis, the National Institutes of Health in the U.S. purchased Bioalma’s software to organize its library and facilitate searches for NIH researchers.

But Bioalma engineers realized the importance of allowing all scientists everywhere, around the world, access to an easily searchable database of scientific papers. So they simplified the program from the one that had been purchased by companies and research institutions. In February 2009 they released novoseek.com as an free online tool available to everyone.

Bioalma alerted the research community around the world, and within only the first month, searches from computers around the world, particularly in the U.S., skyrocketed. Scientists can search for “flu” or “influenza,” and the system understands both terms and searches through all the papers published. Responses may be filtered by symptoms, by treatments, or by all the genes and proteins that could have a relationship to the flu. In addition to published papers, in March they added U.S. grants, and by the end of 2009, the system will include patents.

“We think by the end of the year we’ll be the world leader for a biomedicine search system,” says del Castillo. They
hope to be able to monetize the system through advertisements, as they will regularly reach a broad spectrum of theinternational scientific community.