Precision medicine is having a moment.

All over the news these days, this evolving field takes into account individual variability in genes, environment, and lifestyle, promising exciting new drugs and diagnostics.

Dr. Atul Butte, founding director of the newly-established Institute of Computational Health Sciences at the University of California, San Francisco, spoke on “Data-Driven Precision Medicine” as part of the NLM Informatics Lecture Series on June 3, 2015. A long-time NLM grantee, Dr. Butte has received numerous honors, including being named a White House Open Science Champion of Change in 2013 for promoting science through publicly available data. He and his students have created multiple biotechnology start-up companies in the Bay Area.

The Lecture in Brief

There’s an urgent need to translate genome-era discoveries into clinical usefulness, but the difficulties in making bench-to-bedside translations have been well described. The emerging field of translational bioinformatics may help. Dr. Butte’s lab builds and applies computational tools to convert hundreds of trillions of points of molecular, clinical, and epidemiological data collected by researchers and clinicians worldwide over the past decade—now commonly known as “big data”—into new diagnostics, therapeutics, and insights into diseases both rare and common. A bioinformatician and pediatric endocrinologist, Dr, Butte highlights how publicly available molecular measurements are used to find new uses for drugs; also, interestingly, how the next generation of biotech companies might even start in your garage.

Nine Butte Bytes

Dr. Atul Butte demonstrates the easy access to and portability of “big data” today.

1. The world is experiencing a data deluge. The average person generates two zettabytes (10²¹ bytes) of information annually, and NASA has so many satellites collecting so many pictures, there aren’t enough astronomers to analyze them.
2. Data is there, waiting for you! Big data does indeed come from little packages—microchips. In the world of genomic data, over 1.6 million such chips (from over 2,000 labs) are now freely available through ArrayExpress, a public repository, and the number doubles every 2-3 years. Dr. Butte pronounced the research results coming from this rich pile of data “retroactive crowd-sourcing.”
3. “Data sitting still is useless. Data needs to stay in motion. Data needs to get us someplace.” Big data can be used to develop diagnostics, like a blood test that shows whether you have a disease or might be at risk for getting it. Dr. Butte’s lab developed one for acute myeloid leukemia.
4. Start with need, not code, in the lab. In his case, when his wife suffered from preeclampsia, a leading cause of maternal and fetal death, there was no diagnostic for it. Thankfully, she recovered, but Dr. Butte and colleagues forged ahead, locating genetic markers for the condition and starting a biotech company using $2 million in seed money. He laid out a chain of events: Grab public data. Launch a company. Sell that company. “Again and again, create a value for public data—today!”
5. The high cost of traditional drug development is only going to go higher. “The cost of developing a drug is about $1 billion over 10 years. Clearly, that’s not sustainable,” Dr. Butte observed. “Compounding the problem is the fact that several major drugs are coming off patent in 2018 and can then be sold generically,” he continued. With few new drugs coming online from the pharmaceutical companies, the pharmaceutical industry is expected to lose one-quarter trillion dollars in 2018. “Where are the new drugs going to come from?” Butte asked. It will be up to researchers to solve this drug development challenge.
6. Find new applications for old drugs. There have been several accidental discoveries of this. Intended to treat high blood pressure and angina, Viagra turned out to have other benefits. Rather than rely on the random, however, vast datasets can be used to test the impact of drugs on people with and without diseases, in effect establishing a match.com for drugs
7. We can’t just keep writing papers, hoping someone will do what we propose. Researchers need to create the company and do the work themselves. “It should not be verboten or voodoo to found a company,” Butte opined. Science continues out of the start-up; the start-up is closer to patients; and the start-up creates jobs.
8. The cost of a complete human genome analysis is dropping rapidly and may ultimately be a negative number for consumers. It currently runs $1,000-1,500 but, following the model of auto insurance discounts for safe drivers, it may be that health insurance firms will pay to have your genome analyzed, if you are willing to take an earnest look at your profile and take action to prevent or treat various threats to your health. That’s another area ripe for research and discovery.
9. The data is out there, ready to be used, and ready to serve as the basis for the next start-up. Not only can anyone with the knowledge create these companies—they should. Individual and public health can greatly benefit. “You have the same access to public data that I do,” Dr. Butte told his large group of listeners.

This evangelist of open data then sent them out into the world, the possibility of discovery almost palpable in the air.

By Melanie Modlin, NLM in Focus writer

Related Information
The archived lecture: Data-Driven Precision Medicine
A brief video of Dr. Butte talking about the possibilities of precision medicine:  Faces of the Precision Medicine Initiative: Dr. Atul Butte