Scientists Are Building an “AccuWeather” for Germs to Predict Your Risk of Getting the Flu
Applied mathematician Sara del Valle works at the U.S.'s foremost nuclear weapons lab: Los Alamos. Once colloquially called Atomic City, it's a hidden place 45 minutes into the mountains northwest of Santa Fe. Here, engineers developed the first atomic bomb.
Like AccuWeather, an app for disease prediction could help people alter their behavior to live better lives.
Today, Los Alamos still a small science town, though no longer a secret, nor in the business of building new bombs. Instead, it's tasked with, among other things, keeping the stockpile of nuclear weapons safe and stable: not exploding when they're not supposed to (yes, please) and exploding if someone presses that red button (please, no).
Del Valle, though, doesn't work on any of that. Los Alamos is also interested in other kinds of booms—like the explosion of a contagious disease that could take down a city. Predicting (and, ideally, preventing) such epidemics is del Valle's passion. She hopes to develop an app that's like AccuWeather for germs: It would tell you your chance of getting the flu, or dengue or Zika, in your city on a given day. And like AccuWeather, it could help people alter their behavior to live better lives, whether that means staying home on a snowy morning or washing their hands on a sickness-heavy commute.
Sara del Valle of Los Alamos is working to predict and prevent epidemics using data and machine learning.
Since the beginning of del Valle's career, she's been driven by one thing: using data and predictions to help people behave practically around pathogens. As a kid, she'd always been good at math, but when she found out she could use it to capture the tentacular spread of disease, and not just manipulate abstractions, she was hooked.
When she made her way to Los Alamos, she started looking at what people were doing during outbreaks. Using social media like Twitter, Google search data, and Wikipedia, the team started to sift for trends. Were people talking about hygiene, like hand-washing? Or about being sick? Were they Googling information about mosquitoes? Searching Wikipedia for symptoms? And how did those things correlate with the spread of disease?
It was a new, faster way to think about how pathogens propagate in the real world. Usually, there's a 10- to 14-day lag in the U.S. between when doctors tap numbers into spreadsheets and when that information becomes public. By then, the world has moved on, and so has the disease—to other villages, other victims.
"We found there was a correlation between actual flu incidents in a community and the number of searches online and the number of tweets online," says del Valle. That was when she first let herself dream about a real-time forecast, not a 10-days-later backcast. Del Valle's group—computer scientists, mathematicians, statisticians, economists, public health professionals, epidemiologists, satellite analysis experts—has continued to work on the problem ever since their first Twitter parsing, in 2011.
They've had their share of outbreaks to track. Looking back at the 2009 swine flu pandemic, they saw people buying face masks and paying attention to the cleanliness of their hands. "People were talking about whether or not they needed to cancel their vacation," she says, and also whether pork products—which have nothing to do with swine flu—were safe to buy.
At the latest meeting with all the prediction groups, del Valle's flu models took first and second place.
They watched internet conversations during the measles outbreak in California. "There's a lot of online discussion about anti-vax sentiment, and people trying to convince people to vaccinate children and vice versa," she says.
Today, they work on predicting the spread of Zika, Chikungunya, and dengue fever, as well as the plain old flu. And according to the CDC, that latter effort is going well.
Since 2015, the CDC has run the Epidemic Prediction Initiative, a competition in which teams like de Valle's submit weekly predictions of how raging the flu will be in particular locations, along with other ailments occasionally. Michael Johannson is co-founder and leader of the program, which began with the Dengue Forecasting Project. Its goal, he says, was to predict when dengue cases would blow up, when previously an area just had a low-level baseline of sick people. "You'll get this massive epidemic where all of a sudden, instead of 3,000 to 4,000 cases, you have 20,000 cases," he says. "They kind of come out of nowhere."
But the "kind of" is key: The outbreaks surely come out of somewhere and, if scientists applied research and data the right way, they could forecast the upswing and perhaps dodge a bomb before it hit big-time. Questions about how big, when, and where are also key to the flu.
A big part of these projects is the CDC giving the right researchers access to the right information, and the structure to both forecast useful public-health outcomes and to compare how well the models are doing. The extra information has been great for the Los Alamos effort. "We don't have to call departments and beg for data," says del Valle.
When data isn't available, "proxies"—things like symptom searches, tweets about empty offices, satellite images showing a green, wet, mosquito-friendly landscape—are helpful: You don't have to rely on anyone's health department.
At the latest meeting with all the prediction groups, del Valle's flu models took first and second place. But del Valle wants more than weekly numbers on a government website; she wants that weather-app-inspired fortune-teller, incorporating the many diseases you could get today, standing right where you are. "That's our dream," she says.
This plot shows the the correlations between the online data stream, from Wikipedia, and various infectious diseases in different countries. The results of del Valle's predictive models are shown in brown, while the actual number of cases or illness rates are shown in blue.
(Courtesy del Valle)
The goal isn't to turn you into a germophobic agoraphobe. It's to make you more aware when you do go out. "If you know it's going to rain today, you're more likely to bring an umbrella," del Valle says. "When you go on vacation, you always look at the weather and make sure you bring the appropriate clothing. If you do the same thing for diseases, you think, 'There's Zika spreading in Sao Paulo, so maybe I should bring even more mosquito repellent and bring more long sleeves and pants.'"
They're not there yet (don't hold your breath, but do stop touching your mouth). She estimates it's at least a decade away, but advances in machine learning could accelerate that hypothetical timeline. "We're doing baby steps," says del Valle, starting with the flu in the U.S., dengue in Brazil, and other efforts in Colombia, Ecuador, and Canada. "Going from there to forecasting all diseases around the globe is a long way," she says.
But even AccuWeather started small: One man began predicting weather for a utility company, then helping ski resorts optimize their snowmaking. His influence snowballed, and now private forecasting apps, including AccuWeather's, populate phones across the planet. The company's progression hasn't been without controversy—privacy incursions, inaccuracy of long-term forecasts, fights with the government—but it has continued, for better and for worse.
Disease apps, perhaps spun out of a small, unlikely team at a nuclear-weapons lab, could grow and breed in a similar way. And both the controversies and public-health benefits that may someday spin out of them lie in the future, impossible to predict with certainty.
Did Anton the AI find a new treatment for a deadly cancer?
Bile duct cancer is a rare and aggressive form of cancer that is often difficult to diagnose. Patients with advanced forms of the disease have an average life expectancy of less than two years.
Many patients who get cancer in their bile ducts – the tubes that carry digestive fluid from the liver to the small intestine – have mutations in the protein FGFR2, which leads cells to grow uncontrollably. One treatment option is chemotherapy, but it’s toxic to both cancer cells and healthy cells, failing to distinguish between the two. Increasingly, cancer researchers are focusing on biomarker directed therapy, or making drugs that target a particular molecule that causes the disease – FGFR2, in the case of bile duct cancer.
A problem is that in targeting FGFR2, these drugs inadvertently inhibit the FGFR1 protein, which looks almost identical. This causes elevated phosphate levels, which is a sign of kidney damage, so doses are often limited to prevent complications.
In recent years, though, a company called Relay has taken a unique approach to picking out FGFR2, using a powerful supercomputer to simulate how proteins move and change shape. The team, leveraging this AI capability, discovered that FGFR2 and FGFR1 move differently, which enabled them to create a more precise drug.
Preliminary studies have shown robust activity of this drug, called RLY-4008, in FGFR2 altered tumors, especially in bile duct cancer. The drug did not inhibit FGFR1 or cause significant side effects. “RLY-4008 is a prime example of a precision oncology therapeutic with its highly selective and potent targeting of FGFR2 genetic alterations and resistance mutations,” says Lipika Goyal, assistant professor of medicine at Harvard Medical School. She is a principal investigator of Relay’s phase 1-2 clinical trial.
Boosts from AI and a billionaire
Traditional drug design has been very much a case of trial and error, as scientists investigate many molecules to see which ones bind to the intended target and bind less to other targets.
“It’s being done almost blindly, without really being guided by structure, so it fails very often,” says Olivier Elemento, associate director of the Institute for Computational Biomedicine at Cornell. “The issue is that they are not sampling enough molecules to cover some of the chemical space that would be specific to the target of interest and not specific to others.”
Relay’s unique hardware and software allow simulations that could never be achieved through traditional experiments, Elemento says.
Some scientists have tried to use X-rays of crystallized proteins to look at the structure of proteins and design better drugs. But they have failed to account for an important factor: proteins are moving and constantly folding into different shapes.
David Shaw, a hedge fund billionaire, wanted to help improve drug discovery and understood that a key obstacle was that computer models of molecular dynamics were limited; they simulated motion for less than 10 millionths of a second.
In 2001, Shaw set up his own research facility, D.E. Shaw Research, to create a supercomputer that would be specifically designed to simulate protein motion. Seven years later, he succeeded in firing up a supercomputer that can now conduct high speed simulations roughly 100 times faster than others. Called Anton, it has special computer chips to enable this speed, and its software is powered by AI to conduct many simulations.
After creating the supercomputer, Shaw teamed up with leading scientists who were interested in molecular motion, and they founded Relay Therapeutics.
Elemento believes that Relay’s approach is highly beneficial in designing a better drug for bile duct cancer. “Relay Therapeutics has a cutting-edge approach for molecular dynamics that I don’t believe any other companies have, at least not as advanced.” Relay’s unique hardware and software allow simulations that could never be achieved through traditional experiments, Elemento says.
How it works
Relay used both experimental and computational approaches to design RLY-4008. The team started out by taking X-rays of crystallized versions of both their intended target, FGFR2, and the almost identical FGFR1. This enabled them to get a 3D snapshot of each of their structures. They then fed the X-rays into the Anton supercomputer to simulate how the proteins were likely to move.
Anton’s simulations showed that the FGFR1 protein had a flap that moved more frequently than FGFR2. Based on this distinct motion, the team tried to design a compound that would recognize this flap shifting around and bind to FGFR2 while steering away from its more active lookalike.
For that, they went back Anton, using the supercomputer to simulate the behavior of thousands of potential molecules for over a year, looking at what made a particular molecule selective to the target versus another molecule that wasn’t. These insights led them to determine the best compounds to make and test in the lab and, ultimately, they found that RLY-4008 was the most effective.
Promising results so far
Relay began phase 1-2 trials in 2020 and will continue until 2024. Preliminary results showed that, in the 17 patients taking a 70 mg dose of RLY-4008, the drug worked to shrink tumors in 88 percent of patients. This was a significant increase compared to other FGFR inhibitors. For instance, Futibatinib, which recently got FDA approval, had a response rate of only 42 percent.
Across all dose levels, RLY-4008 shrank tumors by 63 percent in 38 patients. In more good news, the drug didn’t elevate their phosphate levels, which suggests that it could be taken without increasing patients’ risk for kidney disease.
“Objectively, this is pretty remarkable,” says Elemento. “In a small patient study, you have a molecule that is able to shrink tumors in such a high fraction of patients. It is unusual to see such good results in a phase 1-2 trial.”
A simulated future
The research team is continuing to use molecular dynamic simulations to develop other new drug, such as one that is being studied in patients with solid tumors and breast cancer.
As for their bile duct cancer drug, RLY-4008, Relay plans by 2024 to have tested it in around 440 patients. “The mature results of the phase 1-2 trial are highly anticipated,” says Goyal, the principal investigator of the trial.
Sameek Roychowdhury, an oncologist and associate professor of internal medicine at Ohio State University, highlights the need for caution. “This has early signs of benefit, but we will look forward to seeing longer term results for benefit and side effect profiles. We need to think a few more steps ahead - these treatments are like the ’Whack-a-Mole game’ where cancer finds a way to become resistant to each subsequent drug.”
“I think the issue is going to be how durable are the responses to the drug and what are the mechanisms of resistance,” says Raymond Wadlow, an oncologist at the Inova Medical Group who specializes in gastrointestinal and haematological cancer. “But the results look promising. It is a much more selective inhibitor of the FGFR protein and less toxic. It’s been an exciting development.”
The Friday Five: How to exercise for cancer prevention
The Friday Five covers five stories in research that you may have missed this week. There are plenty of controversies and troubling ethical issues in science – and we get into many of them in our online magazine – but this news roundup focuses on scientific creativity and progress to give you a therapeutic dose of inspiration headed into the weekend.
Listen on Apple | Listen on Spotify | Listen on Stitcher | Listen on Amazon | Listen on Google
Here are the promising studies covered in this week's Friday Five:
- How to exercise for cancer prevention
- A device that brings relief to back pain
- Ingredients for reducing Alzheimer's risk
- Is the world's oldest disease the fountain of youth?
- Scared of crossing bridges? Your phone can help