Extending data visualization to biology.
Biological research is exploding with genomic, molecular, and chemical data. But analyzing all that information has been difficult and slow, in part because biologists haven’t had good ways to visualize the data—to see it represented graphically on screen so as to help them spot trends and make comparisons. University of Utah computer science professor Miriah Meyer is addressing that problem by developing programs that make it easier for scientists to explore the data they’re generating. For instance, Meyer has built an interactive program that lets researchers compare different organisms’ genomes, which is useful for understanding evolutionary trends. Scientists also benefit when something doesn’t look right on the screen, because that can reveal a mistake in their data that might otherwise take months to uncover.
Although custom visualization tools are used in many other fields, such as economics, computer science, and engineering, they have been surprisingly slow to spread to biology, says Angela DePace, a biologist at Harvard who has collaborated with Meyer. “More often than not, biologists make do with out-of-the-box solutions that are difficult to tailor to their needs,” she says. That tailoring is just what Meyer tries to do. She spends months working with scientists to understand the specifics of their projects—and how a graphical representation can help.
Using fast DNA sequencing for medical tests.
Diagnostic tests being developed by Yemi Adesokan and his company could let physicians quickly and cheaply pinpoint features of a patient’s infection, such as whether it is resistant to certain antibiotics, and prescribe the most effective treatment.
In 2009, as a postdoctoral researcher at Harvard, Adesokan cofounded a startup called Pathogenica with the goal of developing commercial applications of DNA-sequencing technologies. Adesokan, the CEO, expects to create a market for tests that use sequencing to detect the microbes behind infections. To identify these pathogens today, scientists must use expensive DNA tests or grow the microbes from a sample—a slow process that doesn’t work for many bacteria. And both methods often fail to detect small differences in DNA that can have a huge impact on the organism’s virulence and resistance to drugs. Pathogenica’s technology can pick out specific regions of a pathogen’s genome, such as the genes involved in its ability to infect its host, and sequence many of these regions simultaneously. It minimizes the amount of sequencing, so Pathogenica’s approach will be cheaper, faster, and more precise than existing tests, says Adesokan.
Pathogenica’s initial efforts have focused on detecting the microbes that cause urinary-tract infections. Its researchers are also developing tests to analyze how microbe populations change when someone is treated with new antibiotics or antivirals. Because the technology can detect small changes in DNA, it may be able to reveal early on if a population of microbes is developing resistance to a drug.
Computer-assisted genetic engineering.
Synthetic biology offers the prospect of engineering microbes to fight disease or produce biofuels, but designing the necessary DNA instructions is normally an arduous task. With software written by Andrew Phillips, who heads the Biological Computation Group at Microsoft Research in Cambridge, U.K., scientists can simply select the actions they want the microbe’s proteins to perform and get back a corresponding DNA sequence.
The software bridges the gap between the kind of instructions biological designers would like to use—for example, “Convert protein A into protein B”–and the complicated reality inside cells, where countless reactions are taking place in parallel. It can generate multiple DNA sequences, coding for different ways a cell might produce the same desired result; users can then simulate the different possibilities. So far, the range of actions the software can handle is limited, but the Microsoft group has already used it to design live bacteria that change color when exposed to different molecules.
Phillips’s software will reduce the number of time-consuming failures in real cells, says Douglas Densmore, a computer engineer and synthetic biologist at Boston University. It will enable designers to engineer biological systems that have “a greater probability of working consistently and correctly.”
Letting consumers analyze their sleep.
In college at Brown University, Ben Rubin had an odd nighttime ritual. He would hook himself up to an old polysomnography machine, a refrigerator-size device that clinics use to diagnose sleep disorders. He wanted to create a wearable alarm clock that would measure brain activity and wake the user in an optimal phase of light sleep. Before he graduated, Rubin cofounded a company called Zeo, and in 2009 it began selling the first consumer device that detects the user’s phase of sleep.
The $200 device consists of a fabric headband with embedded sensors that pick up electrical signals from the brain and send them throughout the night to a base-station clock next to the bed. In the morning, the clock displays the amount of time spent in light, deep, and REM sleep; the number of awakenings during the night; and a score that incorporates all these values into a single number. Accompanying software provides suggestions to improve sleep quality, and users can further explore their sleep data on Zeo’s website.
Zeo’s sleep monitor is one of a growing number of consumer tools designed to monitor health and fitness, and Rubin has become an advocate for the idea that people can take more control of their health. “These tools give people the ability to do that,” he says. “People don’t value sleep as they should.”
Networking patients to combat chronic diseases.
Each day, thousands of people around the world open an automated e-mail asking, “How do you feel now?” The e-mail’s recipients belong to a social network called PatientsLikeMe, and nearly all have been diagnosed with a life-changing illness such as epilepsy, Parkinson’s, chronic depression, or amyotrophic lateral sclerosis (ALS, also known as Lou Gehrig’s disease). By answering this simple question, they are participating in an ambitious experiment that is challenging conventional ideas about health care and accelerating the pace of medical research.
The primary architect of this experiment is Paul Wicks, a neuropsychologist and the director of research and development at PatientsLikeMe. Wicks joined the company in 2006, two years after it was founded by MIT-trained engineers James and Benjamin Heywood and their friend Jeff Cole. Their original aim was to garner ideas for extending the life of the Heywoods’ brother, Stephen, who had been diagnosed with ALS at age 29.
Stephen Heywood died in 2006, less than a year after PatientsLikeMe launched an online community where ALS patients could quantify and share details of their symptoms. But by then the site had demonstrated its potential to help its members. “We already knew we had something special,” Benjamin Heywood recalls. “We were a well-resourced family with access to all the best doctors and scientists. But it was amazing how much we learned from our members from day one.”
It was Wicks who shepherded the transformation of PatientsLikeMe from an online outpost for ALS patients into a thriving, global, passionately engaged community of people with more than 1,000 serious conditions, including cancer, diabetes, and HIV. The site gives patients access to powerful tools that were formerly available only to physicians and researchers, such as clinical assessment tests and algorithms that can predict, from the progression of the disease and other factors, how long someone is likely to live.
Some medical professionals questioned Wicks’s willingness to put these tools into patients’ hands, and they weren’t as confident as he was that self-reported data had real value in clinical research. Wouldn’t patients who could calculate their own prognosis become discouraged? Isn’t self-reported data more subject to bias than traditional clinical trials? But Wicks’s ideas have proved their value the old-fashioned way: with peer-reviewed studies in prestigious journals.
In April, Wicks and his colleagues countered claims by Italian researchers that a drug called lithium carbonate slows the progression of ALS. After a clinical study based on a small sample of patients showed promise for lithium in 2008, hundreds of PatientsLikeMe members began taking the drug, which can have serious side effects.
From the site’s ALS community, Wicks assembled a sample of patients that was 10 times the size of the original study group, using an algorithm that matched lithium users with nonusers whose condition was progressing at a similar rate. The study revealed that despite many patients’ enthusiasm for the drug, it was having no effect on the progression of their disease. Though the news about lithium was disappointing, it was also valuable. When a patient’s life expectancy is measured in precious months, weeks, and days, being able to rule out an ineffective treatment can focus caregivers’ efforts in more productive directions.
The Right Balance
Wicks’s particular gift, say those who have worked alongside him over the years, is his ability to view medicine from every angle: from the perspective of the researcher who strives to understand a disease, the doctor who is working to treat it, and the patient who is struggling not merely to survive but to express individuality in the face of escalating physical challenges. “Paul has a unique ability to dive into the literature, learn about a disease, connect with the experts, talk with them at their level, assimilate that information, and then integrate it,” says Benjamin Heywood. “Getting the right balance between the patient perspective, the clinical perspective, and the research perspective is key for us.”
It’s an ability that Wicks has been developing since his teens. After being inspired to study psychology by books like Oliver Sacks’s The Man Who Mistook His Wife for a Hat, he began working as a summer tutor for autistic children. “Talking to the mothers of these kids as they fought for things like the right of their child to have a special-needs education was amazing,” he says. These conversations helped Wicks, who was then only 17, learn how to speak effectively to people facing significant health issues and help them avoid being exploited by quacks.
By 2002, Wicks was working on his PhD at the Institute of Psychiatry at King’s College London. When an online message board for ALS patients called Build lost its funding, he offered to moderate it as an unpaid volunteer, becoming a trusted intermediary between the Build community and doctors at King’s College Hospital. “Some of our members would get angry and ask questions like ‘Why do we need placebos in clinical trials?’ So I would go and ask the head of the clinical trials unit, he would explain it to me, and I would explain it to them,” he says, stressing that “most of the benefit came out of the patients’ interactions with each other.” Build members swapped tips on using assistive technologies and offered support to those facing grueling medical procedures such as experimental stem-cell therapy.
Meanwhile, Wicks was driving hundreds of miles each week to make neuropsychological assessments of ALS patients in their homes. Many of them, he saw, were adapting to the challenges of their disease with creative tactics such as switching to electric toothbrushes when they lost their manual dexterity, or rigging up remote-control systems to accomplish tasks they could no longer perform themselves. “My favorite TV show as a kid was MacGyver [a show about an adventurer who would escape peril using items like duct tape and a Swiss Army knife], and these patients had MacGyvered their own solutions,” he says. When Wicks heard about PatientsLikeMe from Build members, he saw an opportunity to aggregate and disseminate these solutions on a large scale.
PatientsLikeMe now has more than 110,000 members, from whom it solicits information about all aspects of their condition: their medical history, their course of treatment, the medications they’re taking (including those they’re using “off label,” in unapproved applications), the adaptations they’re making, and the impact of their diagnosis on their lives and those of their loved ones. (The company makes money selling aggregated member data to pharmaceutical companies like Novartis and Sanofi-Aventis.) Instead of encouraging its members to be guarded and private about the day-to-day realities of coping with serious illness, PatientsLikeMe advocates complete openness. Members can remain anonymous using avatars and pseudonyms, but many choose not to.
Very personal profile: PatientsLikeMe users can track and share daily updates on many details of their condition, including their overall mood, the doses of medication they are taking, and the severity of their symptoms.
Gary Rogers, a 60-year-old kitchen designer who lives in North Carolina, joined PatientsLikeMe in 2010, seeking support after his diagnosis of multiple sclerosis and polymyalgia rheumatica, a painful inflammatory disorder. Before he became a member, Rogers doubted that participating in a social network would be very useful, but the site has helped him manage his health and stay connected with others.
“Each of us has a different set of symptoms to adjust to,” he says. “Sharing these, as well as current treatment options, helps us decide which path to take and what we need to discuss with our doctors. Without PatientsLikeMe, I would feel alone and isolated, looking for information on impersonal research sites.” He adds, “Most of us don’t get out much, and being able to ‘visit’ with someone else helps relieve the cabin fever. The most important thing for me has been the friends I’ve made on the site and the emotional support they provide.”
Wicks is pleased with what he’s accomplished in five years, but he’s still haunted by the social cost of incurable illness. “We need more people not being held back—not being dragged down, unable to participate because of health problems,” he says. Defeating ancient afflictions like cancer and heart disease would create another challenge for society, he acknowledges: “the problem of everyone living to be 150.” But that, he says, is a problem he would love to have.
Reprogramming stem cells to repair blood vessels.
Injury and disease can damage blood vessels. But Fan Yang, a Stanford professor of bioengineering and orthopedic surgery, has developed a way to persuade the body to repair them.
In her technique, stem cells are reprogrammed in the lab to produce a protein that stimulates the growth of blood vessels. Then the cells are injected into diseased areas of the body. Previous attempts to use this approach ran into problems because researchers relied on viruses to transport the protein-producing genetic instructions into the stem cells. Instead, Yang has made a biodegradable polymer that binds weakly to strands of DNA, clumping together to form nanoparticles that penetrate the stem cells and release the desired instructions. Because these polymers degrade naturally after use, the treatment is potentially safer than viral methods.
Yang believes that eventually the technology could be used to treat the damage caused by heart attacks, strokes, and diabetic ulcers. She’s now collaborating with Stanford surgeons and further improving the nanoparticles, but she estimates that it will be five to 10 years before the therapies move “from bench to bedside.”