Pharma R&D Today

Ideas and Insight supporting all stages of Drug Discovery & Development

Select category

Search this blog

Embracing and Releasing the Power of Data

Posted on May 24th, 2016 by Jaqui Hodgkinson in Pharma R&D

Big data have huge potential in pharmaceutical research, especially now that the technology to manage, mine and assess the vast amount of available information is catching up with the reality of researchers’ needs. Where it once would have taken many months to effectively trawl through the vast amount of literature and experimental data produced annually, relevant answers are often just a quick search away.

Navigating the deluge of data can still seem to be a daunting task. Every year, PubMed adds more than 25 million references and over 1 million new biologically-focused articles are published. Medical research facilities are producing more data on patients’ co-existing conditions and genetic makeup. In silico modeling of diseases and compound–target interactions reveal new insights into underlying mechanisms and molecular relationships.

The idea that someone might drown in a sea of references and information is still there, especially if their only experience of search engines are those that return long lists of citations or if they always have to query multiple internal and commercial databases to find one answer. The reality is that trying to find a single relevant data point with such a simple search engine and siloed information structure would be akin to navigating the Atlantic Ocean without a compass: you might be lucky and arrive at your destination, but you will certainly take a long time—and take quite a few wrong turns along the way!

Fortunately, research solutions have come so far that there is no reason to use such approaches. Scientists can handle immense data sets thanks to the tools that have evolved in response to their demands. It is possible to integrate information from multiple sources—including internal databases—in a single repository. Normalization procedures facilitate comparison of data, regardless of its source, and applying relevant vocabularies and taxonomies lends structure, making it easy to discover whatever answers are needed.

Searching is then fast and easy thanks to refined search algorithms. Scientists use interfaces to form the queries that are most appropriate to their needs, retrieving extracted data from the database. Even when dealing with full text, answers can be obtained. Modern text-mining tools comb taxonomically organized repositories of full-text literature for relevant content. The answers from all searches are delivered it in a form that is ready for in silico modeling and other analyses, all of which should be possible within the same tool environment.

When the researcher has the desired information about, for example, the pathways involved in a disease, the substructures and physicochemical of a drug-like compound, or the adverse events experienced with a similar compound, the real decisions for R&D can begin. The power of big data is its ability to drive decisions. In pharmaceutical research, big data means, among other things, finding relevant associations between genes, proteins and chemicals.

Embracing the modern research solutions that deliver answers quickly is essential to the reinvigoration and success of the pharmaceutical industry. Rather than feeling daunted by a deluge of data, we encourage scientists we work with to utilize the power of the modern ways of managing, mining and analyzing their vast information resources.