Mark thinks big…data
There is no doubt we’re living in the era of “big data.” We have grown accustomed to having information readily available whenever we want it.
As a result, the human race (yes, we) have created more data in the past two years than in our entire history on this earth. But what do we do with the data once we have it?
Data could be anything from search queries, email messages, photos and videos and the more you have of it, the bigger the data set. Big data is essentially extremely large data sets that can be analyzed to reveal trends and patterns. So how is all of this information making our lives better?
As a company, we generate a tractor trailer load of data reports for each new medicine we file for approval. That means there could be a “Eureka!” discovery buried somewhere in piles of unanalyzed, unused data – the key to discovering new and better medicines, faster and maybe even cheaper is within reach and we don’t even know it. It’s a problem no one in the life sciences has solved yet.
Mark Ramsey, senior vice president and chief R&D data officer in Richardson, Texas, would like to change that.
“Our data was in silos, so it was difficult to take what we learned across the R&D pipeline and build on it,” said Mark. “To gain the new levels of efficiency and insight we needed to reduce our costs and speed development, we had to create a platform that would ingest all unstructured and structured R&D data, and deliver greater analytic capabilities.”
With that in mind, Mark pulled in some partners you don’t always find teaming with the chemists and biologists -- data wranglers, data mungers, programmers, mathematicians, statisticians and masters of analysis. They gathered more than 10 years of biological data that was stored in different formats and different places and for more than a year, teams on three continents aggregated, integrated, curated and summarized data.
The result: our scientists can now quickly search more than 10 million rows of data and, with just a few clicks, they can pull out the answer to questions such as what are all the experimental results of compounds tested against one key biological target? What is the best compound to progress to drug development? What’s the ratio of men to women in our clinical trials? (55:45 male:female across 101,000+ patients, if you’re curious!)
But we haven’t stopped there. Our partnership with BioBank, for example, has opened access to health information from 500,000 volunteers in the UK. Integrating this data with our own can help us examine whether there’s a link between genomic data (DNA) and various health conditions.
So far, we have looked for DNA variants that may be associated with a person’s body mass index. We’re also looking for variants associated with asthma, heart attack history and cholesterol levels.
We are the first to do this type of data mining on this scale.
So, back to the original question -- how does this help make our lives better?
In it simplest terms we want to make better decisions about what medicines to pursue and how to progress them. Harnessing this large volume of data delivers unprecedented information for our scientists to use.
And we’re moving faster. Previously, it could take more than a month to map and conform clinical trial data from a single study for analysis. Now researchers can process roughly 120 studies per month.
“We’ve created a platform that provides our scientists with insights that can shorten delivery timelines, reduce costs, expand reach, increase safety, and, in the end, help to save lives,” said Mark.