CWWeekly presents this biweekly feature as a spotlight on issues that executives in clinical research face. This week, writer Suz Redfearn spoke with Daniel Levitt, CEO of Bioz, a new search engine built for life science experimentation that mines research pages, including NIH data, to make the knowledge available to researchers.
Q: The technology to access information about previous research has lagged in the life science sector. Why is that?
A: It really comes down to two main things.
Information about previous research can be found in life science articles, and to access this information, researchers need to manually read through dense and complex text to find relevant insights about experiments. This has always been a cumbersome process, and unfortunately the technology didn’t yet exist to help researchers sift through the hundreds of millions of article pages, with a new life science article being published every 10 seconds.
Natural language processing—NLP— technology previously was not advanced enough to automate the accurate extraction and structuring of experimentation data from within articles. Moreover, the lack of standardization in how researchers write articles set the bar very high, and required very advanced NLP technology.
But now it’s here.
The second reason why the life science sector has lagged behind in the ability to access information on previous research is that the interaction between life science researchers and information technology experts has been difficult. On one side, researchers are working on either drug discovery or basic science, and are not focused on developing software solutions. On the other side, IT experts who are used to developing software applications for consumers and businesses are not focused on life science.
To IT experts, the whole life science sector is this other world, this closed box; as soon as they hear the words “reagent” or “assay,” they think that a Ph.D. is needed to work in this domain. The two sides haven’t really taken the time to understand each other. It’s a disconnect. Researchers have been waiting for IT experts to develop software to help them easily search through previous research before getting started, and before wasting money and months or more of time on experiments that fail due to products not performing as expected.
Q: How has the gap in technology effected researchers and their experiments?
A: To perform research, you come up with a hypothesis, then plan and execute an experiment. It’s time consuming, and anything that can speed up that process while reducing mistakes is of great value to the researcher. Realizing months or years after starting a study that errors were made with the materials used is devastating. There’s the money aspect, and the human capital cost, then of course the cost in buying reagents, tools and consumables that don’t turn out as expected.
There is a very big industry in which $80 billion a year is spent on life science tools, human capital and time. So much of that could be saved if a researcher could, for example, identify antibodies ahead of time based on other people’s research instead of having to do that work all over again.
One researcher I spoke with from Stanford said it took him six months just to identify a couple of antibodies to be used in his experiment. This was because when he did his research, there was no mechanism to find, based on the research of others in the field, what was most likely to work successfully for that type of experiment. The only way researchers can get to that today is by manually reading paper after paper. It’s a huge problem.
Q: Where do you see all of this in five years? Ten years?
A: There are about 10,000 companies out there that sell life science tools, some of which sell great products and some that do not. By algorithmically rating products based on their quality, we’ll start to see the vendors that sell the good products float to the top, and the field will get cleaned up. Grant money from the NIH will start to be used more wisely, and that $80 billion can be spent on products that will work. That will all translate into faster, more effective drug discovery and better basic research into finding cures for cancer, Alzheimer’s, Parkinson’s and other diseases. It will impact everything we see being done today in the life science sector.
I liken it to the changes we saw in the travel industry when suddenly we were able to explore travel options instantaneously. That lead to some disruption in that field, but it was good disruption. Now we have the same process taking place in the life science sector, with researchers now able to explore experimentation options instantaneously.
They have been waiting for this.
This article was reprinted from Volume 20, Issue 31, of CWWeekly, a leading clinical research industry newsletter providing expanded analysis on breaking news, study leads, trial results and more. Subscribe »